Jump to:

Product Design

The Inverse Relationship Between Tobacco Use and Body Weight

Date: 1990
Length: 44 pages
509757154-509757197
Jump To Images
snapshot_rjr 509757154-509757197

Abstract

Reviews evidence supporting the relationship between tobacco use and body weight regulation. Presents nicotine research supportive of this concept, and relates nicotine to drugs of addiction. Tabulates comparative body weight studies of: smokers, non-smokers and ex-smokers, and draws causal conclusions that "nicotine is responsible for the changes in body weight in human smokers." Proposes tobacco use alters: energy intake, energy expenditure, or both, thus influencing weight. Postulates taste perceptions and food preferences influence food intake, while nicotine influences hormones that regulate body weight. Compares nicotine to other addictive drugs, and underscores the correlation among tobacco and alcohol use.

Fields

Notes

Appears to be a book chapter.

Author
Grunberg, Neil, Ph.D. (a scientific editor of the 1988 Surgeon General's report)
Plaintiff
Hypothesis
Behavior Targeting
Cigarette's effect of enhancing/mitigating specific behaviors
Health effects
Design changes which have measurably altered health effects of cigarette smoke, both for smokers and nonsmokers.
Measuring human intake
Development of scientifically valid procedures for measuring tar and nicotine levels that more accurately reflect human intake.
Measuring human smoking behavior
Measuring the effects of changes in human smoking behavior on intake of nicotine and smoke constituents.
Nicotine transport, transfer, and uptake
Design changes which alter nicotine delivery or effect how the product causes and maintains dependence, including transfer of nicotine from tobacco to smoke, and uptake into the body.
Smoking psychology and behavior
Keyword
Addiction (Dependence)
Adrenal effects
Animal testing
Behavioral effects (Behavioral pharmacology)
Addiction behavior, withdrawal, and measured nicotine effects
Benefits of Smoking
Body weight regulation (Weight loss, diet)
Brain activity
Cardiovascular system (Heart)
Central nervous system (CNS)
Daily intake
Diuretic (Weight-loss)
Dose-response
Drug effects
Electrophysiological
Ex-smokers
Habituation
Human testing
Inhalation (Smoke inhalation)
Inner need
Lower respiratory tract (Lungs, bronchial tubes)
Lung cancer
Neuropharmacology (Electrophysiology)
Receptor, brain, and CNS effects (EEG, trigeminal response, etc.)
Nicotine delivery (Smoke nicotine or nicotine yield)
Physiological effects
Psychological effects (Experimental psychology)
Perception patterns, inhalation patterns, and effect on delivery
Puffing behavior (Human puff parameters)
Quitters/ Quitting
Self-administration
Sensory response
Smoker behavior (Human smoking behavior)
Puff parameters, daily intake, etc.
Smoking and Health
Social psychology
Coping/stress management, image, and personality
Stimulant
Tumorigenic
Upper respiratory tract (Mouth, throat)
Younger adult smokers
Smoke Constituent
Nicotine
Nicotine salts
Named Organization
@u_health_s
United States Department of Defense
Subject
@addiction_effects
Behavioral Effects (Effects)
Cancer (Health Effects)
Cardiovascular Effects (Health Effects)
CNS/Brain (Effects)
Effects—Smoking Behavior (Effects)
health effects
Nutrition/Weight (Health Effects)
Respiratory Effects (Health Effects)
Smoke Nicotine (Measures)
Test/Animal Subject (Testing)
Transfer to Smoke (Measures)
Brand
LUCKY STRIKES
*VIRGINA SLIMS

Document Images

Text Control

Highlight Text:

OCR Text Alignment:

Image Control

Image Rotation:

Image Size:

Page 1: rcy63d00
;a _ _Q ~,,, G ,/ 7 d ~'"' ti lUi3 LXJ5t3 P 273 9 LXfi5 8 ~fP 4 C r7~0 RES ADV ALCOtiUL UilU& /0 ~ , ,' The Inverse Relationship between Tobacco Use and Body Weight / l f~EIL E. GRUNBERG 1. INTRODUCTION AnA Cigarette smoking in the United States is responsible for one-third of cardiovascular diseases, one-third of cancer deaths, a large number of respiratory diseases, and a substantial proportion of deaths by fire (USDHEW, 1979; USDHHS, 1982, 1983, 1984). Other forms of tobacco use (e.g., smokeless tobacco) are responsible for addi- tional cases of neck and head cancer (USDHHS, 1986; NIH Consensus Conference, 1986). In fact, more than twice as many deaths are caused by tobacco products in this country than by all other addictive drugs combined, including alcohol (USDHHS, 1988). Despite the well-known and extensively documented health risks of cigarette smoking and tobacco use, 51.1 million Americans smoke cigarettes and additional millions consume other forms of tobacco (USDHHS, 1988). Improved understanding of why people use tobacco products and are reluctant to give them up is important in order to design more effective cessation and prevention strategies. Many smokers report that control of body weight is a major reason why they smoke and refuse to quit (Charlton, 1984; Klesges and Klesges, 1988; Page, 1983). This chapter reviews evidence that there is an inverse relationship between cigarette smoking and body weight and discusses the role of nicotine in this relationship. Possible reasons for the inverse relationships between smoking and body weight, and nicotine and body weight, are presented, and studies that examine each of these reasons are reviewed. Mechanisms that may underlie the reasons for the nicotine/body weight relationship are discussed. Finally, the data and postulates linking nicotine and body weight are consid- ered with respect to other pharmacological agents of addiction. z NEIL E. GRUNBERC Iv]fdcaI Psycho gy Department, Uniformed Services University of the Health Sciences, Bethesda, Maryland 20814-4799. The opinions or assertions contained herein are the private ones of the author and are not to be construed as official or reflecting the views of the l)epartment of Defense or the Uniformcd Services University of the Health Sciences. csti 11523T'1)~ LCIF'YKli~tiT n~ pLr.;JJP1 PUt3l~ Y >rl rC t((iC 90 )ttP PdY r - ~ ~ 273 " ~~~L
Page 2: rcy63d00
274 NEIL E. GRUNBERG . 2. CIGARETTE SMOKING AND BODY WEIGHT The generalization that cigarette smoking helps to control body weight is supported by two types of data: (1) cross-sectional between-subjects comparisons of the body weights of smokers and nonsmokers; and (2) longitudinal within-subject assessment of the body weights of smokers who quit smoking compared to nonsmokers. These two data bases are reviewed. Between-Subjects Comparisons of the Body Weights of Smokers, Nonsmokers, and Ex-smokers Studies comparing the body weights of smokers and nonsmokers date back 100 years. There are several reports published in the 19th century that tobacco use stunts growth (Otis, 1884; Kitchen, 1889; Seaver, 1897). Between 1900 and 1970, many studies reported that tobacco use was associated with lower body weights (Anderson, 1914; Ashford et al., 1961; Blackburn et al., 1960; Bogen, 1929; Fink, 1921; Heath, 1958; Higgins and Kjelsberg, 1967; Holt, 1921; Jenkins et al., 1968; Lickint, 1933; Pemberton and Macleod, 1956; Taylor, 1910). However, some studies conducted dur- ing this period reported no significant weight differences between smokers and non- smokers (Anderson, 1914; Diehl, 1929; Hadley, 1941; Holt, 1922-1923; Peters and Ferris, 1967; Shah et al., 1959; Short et al., 1939; Turley and Harrison, 1932). Grunberg (1980, 1986a) pointed out that most of the null findings regarding smok- ing and body weight were based on studies of adolescents and young adult males. He suggested that the limited exposure of these subjects to tobacco may be partially respon- sible for the lack of a smoking/body weight relationship. Also, the high metabolic rates that naturally occur for young people may have masked or offset any effects of tobacco on body weight. Possibly, the sex of the subjects contributed to the weak findings as well. Studies that included a large age range of subjects report a consistent inverse relationship between cigarette smoking and body weight. Moreover, examination of subjects. by age group supports Grunberg's (1980, 1986a) suggestion that the null findings reported by some investigators are attributable to the youth of their subjects. For example, Higgins and Kjelsberg (1967) compared the body weights of 5020 smok- ers and nonsmokers who ranged in age from 16 to 79 years. Over all ages smokers weighed significantly less (roughly 8 lb less) than did nonsmokers. However, the 16- to 19-year-old smokers had body weights similar to their nonsmoking counterparts. Kopczynski (1972) studied 1245 smokers and 1814 nonsmokers and similarly reported that, except for 19- to 20-year-old males, smokers weighed less than comparably aged nonsmokers. A recent review (USDHHS, 1988) of cross-sectional studies on smoking and body weight published between 1971 and 1987 revealed the following (see Table 1): (1) 25 of the 28 (i.e., 89%) studies found that smokers weighed less than nonsmokers; (2) one study reported this relationship for women but not for men (Sutherland et al., 1980); (3) one study reported this relationship for 45- to 49-year-old men but not for 40- to 44-year- old men (Hjermann ct al., 1976); (4) one study did not find a body weight difference (Waller and Brooks, 1972).
Page 3: rcy63d00
Table 1. Cross-Sectional Evaluations of Smoking and Body Weight- Study Design and sample Albanes et al. 12,103 men and women, (1987) NHANES II Survey Major results Moderator variables Limitations Smokers weighed 5.95 lb less than nonsmokers, controlled for age, sex; smokers taller and leaner than nonsmokers, based on skinfold Age: current smokers gained less after age 25 than either non- smokers or ex-smokers Smoking duration: body mass in- dex decreased with smoking duration increase Smoking rate: moderate smokers leaner than low or high rate smokers Smoking self-report Andrews and All 18,631 pregnant women, Across all heights, smoking Pregnant women only; birth sur- McGarry Cardiff, Wales, 1965-1968 mothers lighter than non- vey record data; actual weight (1972) smokers changes not presented Biener (1981) 274 (174 men, 100 women) ex- 49% women, 39% men gained Retrospective postcessation gain smokers, worksite setting weight following cessation; self-report; no nonsmoker con- quitter approximate average trol group gain: women 11 lb, men 15 lb Blair et al. 183 white male, 284 white Smokers 2.64-7.5 lb lighter Small sample size; white office (1980) female insurance company than nonsmokers, 0.88-15.21 workers only employees; average age 34 lb lighter than ex-smokers; smaller skinfolds for smokers of both sexes than nonsmokers Bjelke (1971) 8,638 male, 10,331 female re- Used "bulk index" Smoking rate: not related to Self-report by mail; no weights, spondents, mail survey. Nor- (weight/height2); both sexes weight no statistical analyses way general population current smokers less bulky Age: older respondents greater presented "systematic sample" than quitters and never smoker/nonsmoker bulk smokers differences Sex: women greater smoker/nonsmoker bulk dif- ferences (continued)
Page 4: rcy63d00
Study Design and sample Major results Moderator variables Fehily et al. 211 nonsmoking, 282 smoking Smokers weighed 7.5-10.3 lb (1984) men, aged 45-59, heart dis- less than nonsmokers, 6.6-9.4 ease study lb less than ex-smokers; pipe/cigar smokers weighed 2.4 lb more than nonsmokers; weight/height2 index results similar Fisher and Gor- 15% random sample, 10 U.S., Men: smoking nondrinkers don (1985) Canadian clinics; 2269 male, weighed 6.6 lb less than non- 2105 female whites, aged 20- smoking nondrinkers; smoking 59, LRC Prevalence Study drinkers weighed 2.2 lb less than nonsmoking drinkers Women: smoking nondrinkers weighed 2.2 lb less than non- smoking nondrinkers; smoking drinkers weighed 4.4 lb less than nonsmoking drinkers Friedman et al. 38 smoking-discordant mono- Smokers weighed 5.07 lb less (1981) zygotic twin pairs, average than nonsmokers age 40 years Garn et al. 17,649 pregnant women, na- (1978) tional health survey Smoking mothers prepregnancy weight less than nonsmoking mothers; difference: whites Limitations Small, all white, restricted sam- ple; smoking self-report All white population; smoking self-report Self-report by mail; small re- stricted sample SES and race: no smok- Pregnant women only; self-re- ing/weight relationship influ- ence Sample size, weights not given; no statistical evaluation 2.43 Ib, blacks 3.53 lb Garrison et al. Framingham study participants; Nonsmokers 55% of highest (1983) assessed 1949-1952 weight group; smokers 80% of lowest weight group Goldbourt and 10,059 male government work- Current smokers 1/4 inch taller, Medalie ers, aged 40-65 2.36 lb less than nonsmokers; (1977) ex-smokers in between; leaner skinfolds for smokers than ex- smokers and nonsmokers Gyntelberg and 5249 employed men, aged 40- Nondrinking smokers 1.5 per- LSIL SL60S ports Limited age range, employment group; smoking self-report All-male sample, one city;
Page 5: rcy63d00
Meyer (1974) 59, Denmark Hjermann et al. Approximately 18,000 male par- (1976) ticipants, aged 40-49, coto- nary risk factor screening, Oslo Holcomb and 226 manufacturing company Meigs (1972) male hourly employees, aged 55-59 centile points lighter than non- drinking nonsmokers; light drinking smokers 2.9 percen- tile points lighter; heavy drinking smokers 5.9 percen- tile points lighter than drink- ing nonsmokers Aged 45-49 smokers body weight 3.09 lb less than non- smokers; aged 40-44 dif- ference not significant; no group weight/height2 index differences Mild to moderate smokers 14 lb lighter than never-smokcrs, ex-smokers, and heavy smokers Smoking rate: heavy smoker (>20/day) body weights high- er than lighter smoker Age: older smokers (45-49) weighed less than non- smokers; younger smokers (40-44) no effect Smoking rate: heavy smokers (> I pack/day) heavier than lighter smokers, equivalent to nonsmokers Huston and 184 men, British field regiment <10 mm subscapular skinfold Stenson men averaged 22 ciga- (1974) rettes/day; ?15 mm sub- scapular skinfold men averaged 12 cigarettes/day Jacobs and Got- 3291 white men and women, Smokers lighter than never Smoking rate: male moderate tenborg aged 20-59, no cardiovascular smokers and quitters smokers (14-29 ciga- (1981) disease or elevated risk fac- rettes/day) 6.39 lb lighter than tors; randomly selected mid- nonsmokers, 2.65-9.93 lb dle-class suburb census tract lighter than light and heavy blacks smokers; female moderate smokers 5.07 lb lighter than never smokers, 1.54-8.38 lb lighter than heavy smokers Age: moderate/never-smoker weight difference increased with age smoking self-report Smoking self-report; limited age range; one city; all men Smoking self-report; limited age, incomes; all men Limited male sample; smoking self-report; no separate smoker/nonsmoker data Smoking self-report; restricted population
Page 6: rcy63d00
Table 1. (Continued) Study Design and sample Major results Khosla and 10,482 male steel workers, Per weight/height2 index, smok- Lowe (1971) Wales ers lighter than nonsmokers Kittel et al. 8284 male factory workers, Relative weights significantly (1978) Belgium lower for cigarette smokers than never-smokers, ex- smokers, and pipe/cigar smokers Kopczynski 3059 random selectees, pulmo- Nonsmokers heavier than smok- (1972) nary disease study, Poland ers, except 20-year-old men Lincoln (1970) 3220 male household heads, Smokers weighed 3-14 lb less aged 41-70, across United than nonsmokers States Matsuya (1982) 90 telephone employees, Japan Ex-smokers weighed 5.29 lb more than nonsmokers; light smokers 2.87 lb less, heavy smokers 0.44 lb less than ex- smokers Nemery et al. 210 steelworkers, aged 45-55, Smokers weighed 12.13 lb less (1983) _10 years service, Belgium than never-smokers. 14.33 lb less than ex-smokers 65tL SL60S Moderator variables Limitations Smoking rate: heavy smokers Smoking self-report; restricted (>35 cigarettes/day) heavier population than moderate smokers (15- 34) Age: group weight differences increased after age 35 Limited population, risk factor treatment program Sex, age, smoking rate: no Smoking self-report; weights not smoking/weight relationship reported influence SES: smoker/nonsmoker weight Restricted population; men difference increased as income decreased Smoking rate: heavy smokers (?21 cigarettes/day) weighed 4 lb more, moderate smokers (11-20) 4 lb less than all- smoker average Small, nonrepresentative sample; data self-report Restricted population; smoking self-report
Page 7: rcy63d00
Stamford ct al. 164 (56 smokers. 108 non- Smokers weighed 11.96 lb Ics., (1984a) smokers) premenopausal had lower average Quetelet women; smokers: >_20 ciga- Index than nonsmokers rettes/day, ?5 years, inhale Stamford et al. 269 adult men, fitness center Smokers weighed 14.99 lb less, (I984b) screened; smokers: ?20 ciga- had 12% less body fat than rettes/day, ?5 years, inhale nonsmokers Stephens and 15,518 persons aged > 10; ques- Smokers weighed less than non- Pederson tionnaire, anthropometry smokers; female smokers (1983) weighed 1.32 lb more to 5.73 lb less than female non- smokers; men weighed 3.09- 7.7 lb less; smokers averaged 3.445 lb less than nonsmokers Sutherland et al. Random sample, 175 men and Weight/height2 index and skin- Sex: male smokers not signifi- (1980) women, rural town, New Zea- folds significantly higher in cantly leaner than non- land nonsmoking than smoking smokers; smoking women women; higher for nonsmok- lighter than nonsmoking ing men, but not significant women Waller and 2169 health exhibit visitors "Little weight difference" Brooks among current smokers, non- (1972) smokers, and ex-smokers Zeiner- Approximately 15,000 randomly Henriksen selected Norwegians (1976) °From IISDHHS, 1988. Current smokers average and rel- ative weight lower than non- smokers or ex-smokers Small sample size: pre- menopausal women only; data self-report Select sample, exercising men; smoking self-report; heavy smokers White women self-report, smok- ing self-report; no statistical significance tests Smoking self-report; small sam- ple size Smoking self-report; bathroom scale weight; health-conscious population; high % cigar/pipe smokers; no statistical evalua- tions Smoking and weight self-report, questionnaire
Page 8: rcy63d00
Several cross-sectional studies of smoking and body weight have broadly classified daily cigarette consumption (e.g., <15, 15-24, 25-34, 35+) and have compared these self-reported estimates to body weight (Albanes et al., 1987; Hjermann et al., 1976; Holcomb and Meigs, 1972; Jacobs and Gottenborg, 1981; Khosla and Lowe, 1971; Lincoln, 1970; Schoenborn and Benson, 1988; Stephens and Pederson, 1983). Overall, these analyses have yielded inverse relationships between smoking and body weight and clear inverse relationships exist from 0 to a pack or so of cigarettes a day. However, many of these analyses report curvilinear functions, with a trough in body weight at about 20 cigarettes per day. The fact that, in these studies, body weight does not continue to decline with cigarette consumption greater than a pack a day can follow from several different factors. Most conservatively, this curvilinear function potentially raises qualifications on the inverse smoking/body weight relationship. Perhaps "heavy smokers" is a mean- ingful double entendre that identifies a group of sniokers who, like "chippers" (people who do not appear to be addicted to tobacco), are less responsive to the effects of nicotine than are most smokers. This interesting possibility deserves careful considera- tion. Alternatively, there are no qualifications necessary. It is important to remember that these cigarette consumption analyses are cross-sectional and, therefore, compare different people. People who smoke more are usually older, are a smaller percentage of the population and are a less representative subsample, are more likely to be male, tend to consume more alcohol, and are different individuals. Moreover, differences in self- reported amounts of cigarettes consumed per day are questionably related to nicotine intake. Some of these differences could be controlled statistically, but few studies have reported the results with sophisticated multivariate analyses of these variables. Even when they are controlled statistically, the analyses are still cross-sectional and therefore compare different people. To determine the effects of smoking on body weight, there- fore, it is important to examine body weight as smoking behavior changes. In general, cross-sectional studies indicate that smokers weigh less than compara- bly aged, same-sex nonsmokers. It is likely that tobacco use is responsible for the body weight differences between smokers and nonsmokers. However, between-subjects com- parisons are open to several interpretations. Based on these data alone, it remains possible that people who choose to consume cigarettes and other forms of tobacco may weigh less than nonsmokers for some reason other than tobacco consumption. If tobacco use is not the culprit, then cessation of tobacco use should not lead to weight gains. If tobacco use is the key reason for the difference in body weight between smokers and nonsmokers, then cessation of tobacco consumption should be accompanied by weight gains. Within-Subject Comparisons of Body Weight with Changes in Smoking Longitudinal studies of body weights of smokers before and after they quit smoking began appearing in the research literature in the late 1960s (e.g., Peterson et al., 1968, reported that smokers who quit smoking gain weight). Sonic of these studies compared smokers who quit with smokers who kept smoking; some studies included nonsmokers as a control group to allow for changes in body weight with age; and some studies
Page 9: rcy63d00
280 NEIL E. GRUNBERG Several cross-sectional studies of smoking and body weight have broadly classified daily cigarette consumption (e.g., <15, 15-24, 25-34, 35+) and have compared these self-reported estimates to body weight (Albanes et al., 1987; Hjermann et al., 1976; Holcomb and Meigs, 1972; Jacobs and Gottenborg, 1981; Khosla and Lowe, 1971; Lincoln, 1970; Schoenborn and Benson, 1988; Stephens and Pederson, 1983). Overall, these analyses have yielded inverse relationships between smoking and body weight and clear inverse relationships exist from 0 to a pack or so of cigarettes a day. However, many of these analyses report curvilinear functions, with a trough in body weight at about 20 cigarettes per day. The fact that, in these studies, body weight does not continue to decline with cigarette consumption greater than a pack a day can follow from several different factors. Most conservatively, this curvilinear function potentially raises qualifications on the inverse smoking/body weight relationship. Perhaps "heavy smokers" is a mean- ingful double entendre that identifies a group of smokers who, like "chippers" (people who do not appear to be addicted to tobacco), are less responsive to the effects of nicotine than are most smokers. This interesting possibility deserves careful considera- tion. Alternatively, there are no qualifications necessary. It is important to remember that these cigarette consumption analyses are cross-sectional and, therefore, compare different people. People who smoke more are usually older, are a smaller percentage of the population and are a less representative subsample, are more likely to be male, tend to consume more alcohol, and are different individuals. Moreover, differences in self- reported amounts of cigarettes consumed per day are questionably related to nicotine intake. Some of these differences could be controlled statistically, but few studies have reported the results with sophisticated multivariate analyses of these variables. Even when they are controlled statistically, the analyses are still cross-sectional and therefore compare different people. To determine the effects of smoking on body weight, there- fore, it is important to examine body weight as smoking behavior changes. In general, cross-sectional studies indicate that smokers weigh less than compara- bly aged, same-sex nonsmokers. It is likely that tobacco use is responsible for the body weight differences between smokers and nonsmokers. However, between-subjects com- parisons are open to several interpretations. Based on these data alone, it remains possible that people who choose to consume cigarettes and other forms of tobacco may weigh less than nonsmokers for some reason other than tobacco consumption. If tobacco use is not the culprit, then cessation of tobacco use should not lead to weight gains. If tobacco use is the key reason for the difference in body weight between smokers and nonsmokers, then cessation of tobacco consumption should be accompanied by weight gains. Within-Subject Comparisons of Body Weight with Changes in Smoking Longitudinal studies of body weights of smokers before and after they quit smoking began appearing in the research literature in the late 1960s (e.g., Peterson et al., 1968, reported that smokers who quit smoking gain weight). Some of these studies compared smokers who quit with smokers who kept smoking; some studies included nonsmokers as a control group to allow for changes in body weight with age; and some studies
Page 10: rcy63d00
. INVERSE RELATIONSHIP BETWEEN TOBACCO USE AND BODY WEIGHT 281 included ex-smokers who were not smoking over the period of the time studied. A recent review of the 43 longitudinal studies of this type published between 1970 and 1980 (USDHHS, 1988) revealed that 37 of 43 (86%) found that cessation of smoking resulted in weight gains that exceeded weight gains by nonsmokers, and that people who began smoking lost weight compared to nonsmokers who did not begin smoking (see Table 2). Of the six studies that did not report these findings, (1) three studies examined pregnant women (Gormican et al., 1980; Haworth et al., 1980; Rantakallio and Hartikainen- Sorri, 1981); (2) two studies compared subjects who were making broad life-style changes because they were at high risk for cardiovascular diseases (Hickey and Mulca- hy, 1973; Holme et al., 1985); and (3) one study reported too little information to clearly conclude what differences occurred (Kramer, 1982). Based on cross-sectional and longitudinal studies, smokers generally weigh less than nonsmokers and many smokers who quit smoking gain weight. There are indi- viduals, of course, who do not show these effects, possibly because they intentionally diet or exercise to avoid unwanted weight gains. Overall, the inverse relationship between cigarette smoking and weight gain is clear. 3. DRAWING CAUSAL CONCLUSIONS ABOUT SMOKING AND BODY WEIGHT: THE ROLE OF NICOTINE Despite the clarity and consistency of the cross-sectional evidence, this type of data does not allow causal conclusions to be unequivocally drawn. Longitudinal evidence increases confidence in inferring causality, but nothing is as convincing as a true experiment. With respect to the effects of tobacco or its constituents on body weight, experiments with human subjects randomly assigned to smoking or no-smoking condi- tions for days, months, or years are considered unethical and, therefore, are impossible. Animal experiments allow direct manipulation of exposure to tobacco or nicotine and measurement of body weight and consummatory behavior before, during, and after drug administration. In general, animal experiments using different species (guinea pig, dog, hamster, mouse, rabbit, rat) indicate that tobacco or nicotine administration decreases body weight. In the dog, tobacco consumption decreases body weight to the point of complete emaciation (Wright, 1846). Nicotine consumption also reduces body weight of dogs substantially (Favarger, 1906), but subcutaneous (s.c.) injection of nicotine tartrate does not (Esser, 1903). In the guinea pig, exposure to tobacco dust decreases body weight (Kardasevitch and Khais, 1940) and s.c. nicotine decreases body weight (Leschtschins- kaja, 1926). One study reported that chronic nicotine intoxication does not decrease body weight of guinea pigs (Kim, 1936). In the hamster, tobacco smoke reduces body weight (Passey et al., 1959, 1961; Wchner et al., 1976). In the mouse and rabbit, tobacco or nicotine administration reduces body weight (see Larson et al., 1961, for detailed discussions of 33 experi- ments). In the rat, tobacco smoke or nicotine inhibits growth (sec Larson et al., 1961, for a discussion of 12 rat studies published before 1961). More recently, it has been reported that nicotine administration to rats decreases body weight in a dose-dependent manner and that, in addition, cessation of nicotine results in body weight gains (Grun-
Page 11: rcy63d00
Table 2. Longitudinal Evaluations of Smoking and Body Weight" Study Design and sample Blitzer et al. 57,032 women, aged 20-59, (1977) self-help weight loss groups Bosse et al. 1749 adult men, Normative (1980) Aging Study, assessed over 5 years Burse et al. 4 paid volunteers; 11-day base- (1982) line, 21-day quit period, 20- day resumption period Cambien et al. 1097 Paris civil servants, aged (1982) 25-35, screened, randomly assigned, cardiovascular risk factor reduction intervention or control groups; 2-year fol- lowup evaluation Carney and 13 women, 5 men, aged 28-67, Goldberg smoked ?20 cigarettes/day, (1984) ?5 years; 12 male controls; 15 smokers abstained 2 weeks Coates and Li 373 male asbestos-exposed (1983) V9TL SL605 smokers, aged >42; 87% white, mean education 12.8 years. 12 months assessment after cessation effort Major results Moderator variables Limitations Quitters gained 7.0-10.2 lb Smoking rate: weight gain/pre- Smoking and weight self-reports; more than continuing smokers vious smoking rate propor- all women trying to lose tional weight Average 5-year gains: never Age: younger quitters gained Smoking self-reports; all men; smokers 1.81 Ib; former more actual weights not presented smokers 1.87 Ib; current Adiposity: fatter quitters gained smokers 2.00 lb; ex-smokers more who quit 6.34 lb Tar rate: higher pretest tar rate smokers gained most Anxiety: high related to higher gain 3 of 4 gained weight; 1.98 lb Very small sample, paid volun- increase during cessation; 1.76 teers; short-term evaluation lb loss on resumption Treatment group quitters gained Smoking self-report; risk factor 4.85 Ib, control group quitters reduction program participants 7.50 Ib; nonsmokers and no- change smokers gained 1.54 lb in treatment group, 2.2 lb in control Quitters weight change range: Smoking rate/duration: no Smoking self-report; controls -3.09 to +9.0 lb weight change relationship weight changes not reported; Biological variables: weight gain short-term evaluation positively related to lipopro- tein lipase activity in adipose tissue Continuous quitters gained 5.15 lb; continuous smokers gained 0.35 lb Smoking self-report; all male, nonrandom sample
Page 12: rcy63d00
Comstuck and 502 male telephone workers, Stone (1972) aged 40-59, mostly white; 2 assessments 5 years apart Dallosso and 16 (8 men, 8 women) antismok- James (1984) ing clinic participants; mean age, men 47.1, women 35.4; assessed before and 6 weeks after clinic Emont and 125 stop-smoking clinic partici- Cummings pants; pretreatment and I- (1987) month followup assessments Fagerstrom 28 nicotine gum users; abstinent (1987) at 6 months Friedman and Multiphasic health checkup pa- Siegelaub tients; smoked, then quit 12- (1980a) 18 months later (N = 3825) or continued (N = 9392) Garn et al. 6979 women followed through (1978) ?2 pregnancies Garvey et al. 870 white male veterans, aging (1974) study, assessed 4-7 years after initial assessment 5-year followup averaye _ain.: never-smokers 2.43 lb. ex- smokers 5.07 Ib, continuing smokers 2.42 lb; quitters Sm0kim_ r:ur: incR~u.in~, quittrr wcight _ain with heavier pre- quit smoking Smoking .clf-rcport: m~n only 11.24 lb and showed greatest skinfold increases 10 quitters gained 3.00 Ib; 5 continuing smokers lost 0.99 lb Small sample size; smoking self- report; limited followup 76% quitters and slippers (:~5 Nicotine gum: gain/gum use re- Weight gain, smoking self-re- cigarettes/day) averaged 5.8 liable negative correlation for port, confounded by gum use; lb gain heavy smokers; gain not relat- limited followup; incomplete ed to age, sex, marital status, data baseline body weight Infrequent gum users gained Nicotine gum: frequent users Small sample size; measures un- 6.83 Ib, frequent users 1.98 lb gained less weight clear Quitters gained 2-3 lb more Smoking rate: higher initial Smoking self-report; whites only than continuing smokers smoking rate related to greater data weight gain after cessation Higher prepregnancy weights for habitual nonsmokers than ha- bitual smokers: whites 3.4 Ib, blacks 4.1 lb; lower habitual smoker gains between preg- nancies for both races Smoking/weight change signifi- cantly related; recent quitters (<_5 years) gained 4.19 lb more than smokers, non- smokers, former smokers Race: no weight/smoking rela- tionship influence Age: 40-54 quitter weight in- crease greatest Smoking self-reports; restricted population Smoking self-report; exact quit date unknown (continued)
Page 13: rcy63d00
Study Design and sample Glauser et al. 7 male smokers, cessation pro- (1970) gram; assessed preprogram, I month postprogram Gordon et al. 4798 Framingham study partici- (1975) pants: 1498 male smokers, 492 male nonsmokers, 1634 female nonsmokers, 1174 female smokers; examined short-term changes after bien- nial exam 1, long-term effects between biennial exams 4, 10 Gormican et al. 301 pregnancy obstetrics re- (1980) cords, women, aged 17-35 Grinstead 45 subjects (38 women, 7 men), (1981) averagc age 40; evaluated 6 months after cessation treat- ment; saliva thiocyanate ver- ification Gritz et al. (in press) 554 self-quitters (245 men, 309 women), mean age 41.4, 85% Caucasian, 9% black, 4% Asian, 1% Asian-American, 1% Native American; I-year followup Grossarth- 1353 subjects. Yugoslavian vil- Maticek et al. lage of 14,000; every 2d (1983) household oldest member; evaluated 1965-1966, 1969 99TL SL60S Table 2. (Continued) Major results At I-month followup, partici- pants gained 6.4 lb At entry, male smokers weighed 8.0 lb less than nonsmokers; short-term male quitters gained 3.8 Ib, nonsmokers 0.5 Ib, continuing smokers 0.3 lb; new smokers lost 9 lb; too few female quitters to evaluate Smoker, nonsmoker prepregnan- cy weight similar; no last 2 trimester weight gain dif- ference (nonsmokers 24.6 lb, smokers 22.6 lb) During program, 63% subjects averaged 2.88 lb increase, 34% averaged 2.46 lb de- crease; at followup, 37% aver- aged 6.97 lb gain, 43% averaged 3.27 lb loss 35% previous quitters gained, 3% lost; at 1 year, abstainers averaged 6.1 lb gain; relapsers gained 2.71 lb while absti- nent, lost 1.3 lb upon relapse; continuous smokers gained 0.3 lb Smoking reduction/weight in- crease relationship (regression coefficient -0.30) Moderator variables Limitations Smoking self-report; exact quit date unknown Smoking self-report; change analysis, men only Clinic record data; pregnancy weight gain data only Questionnaire, phone interview data Questionnaire, phone interview data Smoking self-report; weights, weight changes not reported
Page 14: rcy63d00
Study Design and sample Jenkins et al. 2318 men (546 never smokers, (1973) 359 previous quitters, 547 light smokers. 866 heavy smokers), aged 39-49, 11 California corporations in Western Collaborative Group Study; changes assessed since age 25; 1960-1969 study Kramer (1982) 175 subjects, commercial cessa- tion program (41 nonpartici- pants or nonlocated, 59 quitters, 75 continuing smok- ers) ?1-year followup Lund-Larsen 12,329 men and women, aged and Tretli 20-49, cardiovascular disease (1982) project; 2 screenings 3 years apart Manley and 39 male, 55 female smokers, Boland cessation program; randomly (1983) assigned, I of 3 4-week treat- ments or attention placebo control; 3-month followup; CO verification Noppa and 1302 Swedish women, aged 38- Bengtsson 60 (1980) L9TL SL605 Major results Moderator variables Limitations Weight loss more likely for light Smoking self-report; weights not and heavy smokers than never presented smokers and quitters 76% nonsmokers. 56% smokers All data self-report; high attri- gained weight; these smokers tion, data loss; presentation mean gain 1.7 lb, these non- incomplete smokers mean gain 3.0 lb Smokers mean and relative Sex: men, women weight Self-report weight less than nonsmokers; change/smoking cessation and female quitters gained 5.95 lb, initiation similar male quitters 7.84 lb; smok- ing-starter men lost 1.98 Ib, women 5.5 lb; smokers, non- smokers little/no change 31% abstinent at followup: ab- stainers averaged 10.93 lb gain, relapsers 6.92 lb Current smokers leaner than nonsmokers; At 6 years, quit- ters gained 7.72 lb; smoking- starters lost 1.54 Ib, non- changers gained 3.09 lb Relapser definition unclear Smoking self-report
Page 15: rcy63d00
Gunn and 89 cessation clinic participants; 43 of 54 (80'7c) quitters gaincd Shapiro all quit at initial evaluation; 3- 2-30 lb (1985) month followup assessment Hall et al. 255 smoker participants (122 Abstainers gained more than Smoking rate: pretest smoking (1986) men, 133 women), 2 smoking smokers at 1 year Icvcl/postcessation weight treatment trials; 6-, 12-month gain positively related followups; biochemical ver- Chronic dieting: chronic diet ification subjects gained most Hatsukami et al. 27 smokers hospitalized 7 days; Quitters gained 1.76 lb in 4 days (1984) 20 subjects smoked 3 days, then quit 4 days; 7 control group subjects smoked throughout Haworth et al. 536 women (234 nonsmokers, (1980) 302 smokers) interviewed last prenatal visit (18%) or within day after delivery (82%) Hickey and 150 men (124 smokers); 6- Mulcahy month, 2-year followups after (1973) myocardial infarction Holme et al. 16,202 Oslo men, aged 40-49, (1985) screening program; 1232 (ele- vated cholesterol or upper quartile coronary risk score) randomly assigned diet/smok- ing intervention or control; 5- year followup Howell (1971) Retrospective, 1121 men, aged 40-54; 15- to 20-year weight gain examinations Hughes and 37 smokers and 19 ex-smokers Hutchinson with pulmonary emphysema (1983) followed ?3 years No smoker/nonsmoker pregnan- cy weight gain difference Quitter, reducer, continuing smoker differences not signifi- cant 17% controls, 24% intervention quit; 1- to 2-year-quitter weight increased more than controls, then decreased to be- low prequit level Smoking, height, weight self-re- port; inadequate statistical evaluation Multiple treatment (e.g., nic- otine gum) participant data in- cluded Small sample size; inpatient en- vironment Smoking self-report; pregnancy weight gain data only Smoking self-report; postmyocardial infarction may motivate healthy behavior Smoking self-report; confounded by high cardiovascular disease risk health intervention; weights not reported Light smokers (<20 ciga- Smoking rate: lower rate related Retrospective report rettes/day) gained 1.9 lb less to less weight gain than heavy smokers, 3.1 lb less than ex-smokers, 3.6 lb less than never-smokers Smokers lost 0.32 Ib/yr, ex- Smoking self-report; pulmonary smokers gained 1.17 Ib/yr; emphysema population significant difference (continued)
Page 16: rcy63d00
Pincherle (1971) 222 upper-class male quitters; followup ? 1 year after first visit Powell and Mc- 29 women, 22 men, 5-day ces- Cann (1981) sation project; 2- and 6-month followup Puddey et al. 66 cessation program volunteers, (1985) pair-matched by age, sex, body mass index; randomly assigned experimental, control groups; 2-week baseline, 6- week treatment, 6-week fol- lowup; thiocyanate, CO ver- ification 28% gained weight; 22% lost At 2 months, 54% gained weight, range 3-20 lb, mean 8.96 lb; all subjects mean 4.69 lb 14 quitters gained 3.97 lb; con- trols 0.44 lb Smoking self-rcport; limited population; incomplete report; no weights presented Smoking self-report; no separate abstainer, smoker data; small sample size Small sample size Rabkin (1984) 40 male, 67 female smokers, assigned to 3 cessation groups; followup 3 weeks post-completion; biochemical verification 67.3% gained weight, average 1.76 Ib; skinfold increase 6.6 mm No age, age at smoking start, rate, relative weight, anxiety correlation to male or female weight change Small sample size; weight self- report Rantakallio and 12,068 pregnant women, n. No smoking/nonsmoking preg- Pregnant women only; smoking Hartikainen- Finland, 1966; 15% smokers nancy weight gain difference self-report; pregnancy weight Sorn (1981) (smoked after 2 months preg- gain data only nant); nonsmoking controls matched for age, parity, place of residence, marital status Rush (1974) 162 low-income urban pregnant Mean pregnancy weight gain Smoking rate: higher rate related Pregnant women only; smoking women, no known medical lower for smokers (0.73 to lower pregnancy weight self-report; pregnancy weight problems, <140 lb preconcep- lb/wk) than nonsmokers (0.90 gain gain data only tion weight; had borne low lb/wk) birthweight infant; randomized controlled nutritional supple- mentation trial (continued)
Page 17: rcy63d00
Study Design and sample Schoenenberger 4421 male MRFIT volunteers, . (1982) aged 35-57, good health but upper 10-15% coronary risk factor score; randomly as- signed intervention or control groups; followup 3 annual vis- its Seltzer (1974) 794 adult white male veterans, average age 45; Normative Aging Study; screened for "high" health level, geo- graphic stability; 214 screened at 5 years Stamford et al. 13 sedentary women, 48-day (1986) successful quitters; 1-year fol- lowup Tuomilehto et 10,940 cardiovascular disease al. (1985) prevention program partici- pants, aged 25-59, random sample, e. Finland; selectees with high blood pressure or hypertensive medicine as- sessed 5 years apart; smoking data from 2264 Vandenbroucke 3091 Netherlands civil servants, et al. (1984) spouses (1583 men, 1508 women), aged 40-65, general health exam; 25-year followup °From USDHHS, 1988. OLtL SL60S Table 2. (Continued) Major results With MRFIT intervention, sig- nificant body weight decrease in smokers (mean 4.6 Ib), nonsmokers (mean 5.8 lb), re- ducers (mean 3.75 lb); quitters average weight change mini- mal (mean 0.55 Ib) At admission, ex-smokers 5.9 lb heavier than nonsmokers, 8.1 lb heavier than current smok- ers; at 5 years, quitters gained 8.0 Ib, continuing smokers 3.5 lb At 48 days; weight increased 4.85 lb; at 1 year, quitter in- creased 18.07 Ib; 3 relapsers reduced weight to baseline levels; per hydrostatic weigh- ing, gain was 96% fat Quitters body mass increased 2.31 lb/m2; starting smokers decreased 1.46 lb/m2 76.6% lean, 65.1% obese men smoked; 22.1% lean, 11.3% obese women smoked Moderator variables Limitations Smoking self-report; confounded by risk factor reduction pro- gram participation; restricted population White veterans; smoking self-re- port Small female sample; smoking self-report Smoking self-report; hyperten- sives Smoking self-report; restricted population
Page 18: rcy63d00
INVERSE RELATIONSHIP BETWEEN TOBACCO USE AND BODY WEIGHT 289 berg, 1982; Grunberg et al., 1984, 1986, 1987, 1988a; Grunberg and Bowen, 1985; Levin et al., 1987; McNair and Bryson, 1983; Morgan and Ellison, 1987; Schechter and Cook, 1976; Wellman et al., 1986). Recent human studies with nicotine polacrilex chewing gum are consistent with the animal findings. Fagerstrom (1987) reported that smokers who quit smoking and chewed nicotine gum are less likely to gain weight than are exsmokers who did not chew nicotine gum. Emont and Cummings (1987) reported an inverse relationship between number of pieces of nicotine gum chewed and body weight gain by heavy smokers (26 cigarettes/day) who quit smoking. Stitzer and Gross (1988) also found that ex-smokers who chewed gum gain less weight than do exsmokers who do not consume nicotine in any fonn. The animal experiments indicate that smoking and nicotine decrease body weight compared to controls. Animal experiments also indicate that cessation of nicotine in- creases body weight. These effects are dose-dependent; i.e., the greater the dose of nicotine, the greater are the effects. Therefore, nicotine affects body weight and proba- bly is the component of tobacco responsible for body weight changes that accompany changes in tobacco use. The results of the nicotine gum studies are consistent with the interpretation that nicotine is responsible for the changes in body weight in human smokers. 4. EXPLANATIONS FOR THE INVERSE RELATIONSHIP BETWEEN TOBACCO USE AND BODY WEIGHT The inverse relationship between tobacco use and body weight may result from changes in energy intake, changes in energy expenditure, or both. Energy intake in- volves consumption of foods from all available taste categories (including sweet, salty, and bland) and nutrient classes (including carbohydrates, proteins, and fats). Energy expenditure involves behavioral factors (physical activity) and biological factors (e.g., resting metabolic rate, thermogenesis). The evidence for each of these factors is dis- cussed in the following sections. Energy Intake Energy intake is affected by changes in consumption of all foods or changes in consumption of specific foods. Food consumption changes are frequently interpreted to be synonymous with changes in appetite (i.e., hunger for food). However, this assump- tion may or may not be true, especially in humans. Decreased food consumption, for example, can reflect (1) decreased general hunger or appetite; (2) dieting or purposeful restraint; (3) dislike for the food that is available; (4) physical illness that makes it difficult or impossible to cat (e.g., ulcers in the mouth; sore throat; nausea); or (5) mental illness that is accompanied by disinterest in food (e.g., depression). Increased food consumption can reflect (1) increased general hunger or appetite; (2) purposeful overconsumption (e.g., carbohydrate load by athletes before a competitive event; exces- sive consumption to "please" the food preparer); (3) particular preference for the available food; or (4) psychopathological eating pattern, such as bingeing.
Page 19: rcy63d00
290 NEIL E. GRUNBERG With regard to the relationship between tobacco use and body weight, one possible explanation is that smokers eat less than do nonsmokers and that ex-smokers increase their food consumption and thereby gain weight (Birch, 1975). This straightforward explanation for the tobacco/body weight relationship, if true, could reflect (1) effects of tobacco or nicotine on general appetite; (2) a substitution of habits involving the hands and mouth; or (3) a replacement of stimuli for oral gratification (as suggested by McArthur et al., 1958). Alternatively, tobacco use could affect consumption of some (e.g., sweets), but not all, food. If the latter explanation is correct, then studies that do not differentiate among specific foods consumed may obtain different results depending on what foods are available. Energy Intake: Human Studies. Before the 1980s few human studies evaluated the relationship between tobacco and food consumption and these studies reached differ- ent conclusions. Lincoln (1969) evaluated retrospective self-reports of food consump- tion by 885 male smokers and nonsmokers and reported (in a'`letter to the editor") that male smokers consume roughly 350 calories per day more than do nonsmokers. In contrast, Gaudet and Hugh (1969) collected retrospective self-reports from 645 people (including 223 who had changed their smoking habits) and reported no significant changes in eating behavior with changes in smoking habits. S. Schachter and P. D. Nesbitt (personal communication) examined food consumption of smokers in the labora- tory during two 8-hr periods: They were allowed to smoke freely on one day and abstained on the other day. Overall, caloric intake was roughly 33% greater during the no-smoking day. In a short-term laboratory setting, Perlick (1977) found that smokers who were deprived of nicotine (by complete smoking abstinence or by smoking low- nicotine cigarettes) ate twice as many gumdrops as did nondeprived smokers and non- smokers. Based on these studies, smoking either increases food consumption, decreases food consumption, or has no effect on food consumption. Perrin et al. (1961) wrote: "There is a belief that smokers do prefer more savoury foods, and that, with abstinence from smoking, sweeter foods are preferred, and this possibility accounts for the subsequent increase in weight that has been reported" (p. 387). The studies performed before 1979 did not carefully examine the possibility that smoking may have decreased consumption of sonie (e.g., sweet), but not all, foods. The Perlick (1977) study clearly reported increased sweet food consumption (i.e., gum- drops) with nicotine deprivation. Because only gumdrops were available in this study, the increased consumption may also reflect changes in general hunger. Grunberg (1980, 1982) reanalyzed the Schachter and Nesbitt data by taste class and found that consump- tion of sweet and salty foods doubled on the no-smoking days and that consumption of all other foods did not change. These findings suggest that smoking or abstinence from smoking results in changes in consumption of sonic, but not all, foods. Caution must be exercised in interpreting these results because the sweet and salty foods in this study were all easily accessible snack foods while the other foods required sonie preparation (e.g., making sandwiches). Also, without nonsmokers as control subjects, it is unclear whether smoking decreased consumption of specific foods, whether abstinence in- creased consumption of specific foods, or both. In a study of quitting smoking, Gilbert and Pope (1982) studied 19 smokers for 2 days; I day subjects could smoke and I day they could not. On the nonsmoking day subjects ate significantly more. A closer examination of food consumption revealed that
Page 20: rcy63d00
INVERSE RELATIONSHIP BETWEEN TOBACCO USE AND BODY WEIGHT 291 on the nonsmokin eanuts) but I da s subjects ate si nificantl more snacks (candies and g y g y p significantly less at mealtimes. Caloric consumption of snack foods as a proportion of I total daily caloric intake increased significantly on the nonsmoking days, suggesting that abstinence from smoking increased consumption of specific foods. Unfortunately, as in the Perlick (1977) study, the sweet and salty snack foods were readily accessible, which may have contributed to their increased consumption. In addition, there were no non- smokers as control subjects. Therefore, it is impossible to conclude from this study that smoking affects specific or general food consumption. In order to compare the effects of cigarette smoking and abstinence on general and spccific food consumption without the ambiguity of these previous studies, Grunberg (1982) conducted a human laboratory experiment. Male and female smokers (N = 28) and nonsmokers (N = 14) were recruited to participate in a food-tasting study that was run around lunchtime. Smokers were divided into two experimental groups: Half of the subjects were allowed to smoke two cigarettes during a baseline period before tasting the foods; half of the subjects were not allowed to smoke at this time. All subjects filled out questionnaires during this baseline period, and the smoking manipulation was done in such a way to avoid suspicion by subjects of the importance of smoking or not smoking in this experiment. After the baseline period, subjects were escorted to a food-tasting room. Nine foods (three sweet, three salty, three bland) were set out in bite-size pieces i in individual bowls. Subjects were instructed to taste each food and to rate each food on a series of taste judgment scales. Almost as an afterthought, the experimenter explained to subjects that after they rated the foods they could feel free to eat whatever they wanted. The bowls of food were weighed before and after the session to determine exactly how much was eaten of each food. In addition to this measure of food consump- tion, subjects were asked to select three foods to eat again in another part of the study. This selection provided a second index of food preferences. Smokers who were allowed to smoke ate less sweet food than did nonsmokers and than did smokers who abstained from smoking. There were no differences in consump- tion of the nonsweet foods among the three groups. In addition, smokers allowed to smoke were less likely to choose sweet foods to eat again thadwere the other two groups. These results indicate that smoking decreases preference for and consumption of sweet foods in particular (Grunberg, 1982). These results are consistent with those of Schachter and Nesbitt (personal communication) and Perlick (1977). j In an attempt to determine whether these findings generalize to the world, Grun- berg and Morse (1984) compared per capita cigarette consumption in the United States between 1964 and 1977 with per capita consumption of every major foodstuff. These years were chosen for analysis because U.S. per capita consumption of cigarettes during this period was a U-shaped function (a trough occurred from 1969 to 1972 following the fedcral government's intensive and highly publicized antismoking campaign). In effect, this curvilinear function of cigarette consumption acted as an experimental manipula- tion. Although per capita food consumption data could not, of course, be separated for smokers and nonsmokers, Grunberg and Morse (1984) reasoned that if the relationship between specific food consumption and tobacco consumption is a robust phenomenon, then analysis of national food consumption should be consistent with the laboratory findings. As predicted, there was a significant inverse correlation between cigarette and sugar consumption, in particular.
Page 21: rcy63d00
292 NEIL E. GRUNBERG More recently, several investigators have conducted prospective studies of eating behavior in smokers who quit smoking. Hatsukami et al. (1984) carefully measured dietary intake in 27 smokers in a hospital setting. Eating behavior was monitored for 3 days and then 20 of 27 smokers quit smoking for 7 days, during which daily eating behavior was measured for all 27 subjects. The smokers who quit smoking showed significant increases in caloric intake and gained an average of 1.8 lb. Stamford et al. (1986) and Robinson and York (1986) also monitored dietary intake carefully and reported that caloric intake increased in smokers who quite smoking. Dallosso and James (1984) reported an increase in dietary intake after smoking cessation (6.5%), but it was not statistically significant. Unfortunately, these studies did not report food consumption separately by taste classes. Rodin (1987) performed a prospective study and reported changes in consumption by food categories: Smokers who gained weight after quitting smoking increased their sugar consumption. These results supported the conclusions of Grunberg (1982) and Grunberg and Morse (1984) that the inverse rela- tionship between smoking and sweet food consumption may partially determine the inverse relationship between smoking and body weight. Taken together, these studies suggest that smoking decreases food consumption, especially of sweet foods, and that cessation of smoking increases food consumption, especially of sweet foods (compared to smokers who keep smoking). These changes in energy intake probably contribute to the body weight changes. Although the results of the human laboratory experiments and prospective human studies are consistent, cross- sectional surveys report conflicting findings. Based on self-report data, some studies hold that there is no difference in food consumption between smokers and nonsmokers (Albanes et al., 1987; Fehily et al., 1984; Fisher and Gordon, 1985; Matsuya, 1982). Other studies indicate that smokers report eating more than do nonsmokers (Picone ct al., 1982; Stamford et al., 1984a,b). These inconsistent findings may reflect the inval- idity of self-report dietary data. Alternatively, these findings may reflect the fact that these studies do not separate foodstuffs into different categories or types. In addition, it is important to consider that most people alter their eating behavior to help control their body weight. A recent cross-sectional study of smoking and eating that partially broke down responses by food types and that evaluated dietary habits by smoking, social class, age, and sex found that smokers were significantly less likely to eat fruits frequently (sweet foods), to eat breakfast, or to eat brown bread. The smokers were more likely to eat fried foods (Whichelow et al., 1988). This cross-sectional study still relied on self- report data, but it considered important variables (e.g., specific food, age, sex) in the evaluation of the results and reported findings consistent with the laboratory and short- term prospective studies, i.e., smokers and nonsmokers differ in consumption of specif- ic foods, including sweet-tasting foods and carbohydrates. The effects of cigarette smoking on eating behavior may be clearly revealed in the laboratory and in short-term prospective studies, whereas some summary data report no food consumption differences between smokers and nonsmokers because nonsmokers and ex-smokers exert self-control to avoid getting fat. Without objective measures of food consumption, separation by food categories, or measures of dietary restraint, it is difficult to interpret most of the survey data. [The Whichelow et al. (1988) study is an exception.] Moreover, because the laboratory and short-term prospective studies of food intake by smokers do not randomly manipulate smoking and do not report nicotine levels
Page 22: rcy63d00
. INVERSE RELATIONSHIP BETWEEN TOBACCO USE AND BODY WEIGHT 293 in the body, based on these studies alone, it is impossible to reach causal conclusions about the effects of cigarette smoking or nicotine on eating behavior. Energy Intake: Animal Studies. Animal experiments that administer tobacco smoke or nicotine to subjects and measure food consumption provide complementary data to the human studies of smoking and eating. Several experiments conducted before 1980 reported that cigarette smoke or nicotine injection decrease food consumption in rats, humans, and guinea pigs (Averett, 1962; Becker and King, 1966; Munster and Battig, 1975; Evans et al., 1967; Passey et al., 1959, 1961; Hughes et al., 1970). Other studies reported that there were no changes or that the decreases were not significant (Kcrschbaum et al., 1965; Schechter and Cook, 1976). Most of these studies also reported that there were body weight decreases that could not be explained by the decreased food consumption, either because the tinie course of the changes did not match or because the food intake changes were too small (Evans et al., 1967; Schechter and Cook, 1976; Passey et al., 1959, 1961; Hughes et al., 1970). These studies do not conclusively establish whether or not changes in energy intake arc involved in the effects of nicotine on body weight. When food intake decreased, so did body weight. But body weight also decreased at times when food intake did not change. Either something is going on in addition to energy intake or instead of energy intake to account for the inverse relationship between nicotine and body weight. A closer look at these studies also reveals several methodological limitations: (1) None of these studies examined eating behavior or body weight after cessation of nicotine or tobacco smoke; (2) many of the studies that reported decreased food intake measured it for a short pcriod of time immediately after nicotine or tobacco smoke administration; (3) all of these studies provided only one food (a bland laboratory chow). Therefore, the finding of some studies that nicotine or tobacco smoke decreases food intake is unconvincing because the decreased food intake usually occurred immediately after a bolus injection of nicotine or acute exposure to tobacco smoke. It is hardly surprising that immediately after a jolt of nicotine (an extremely toxic substance), food intake may decrease for a short time. In contrast, human sniokers chronically self- administer nicotine and can eat whenever they choose. The conclusion of other studies that food intake does not change much is also open to question because only bland foods were available. In human studies, smoking decreases consumption of sweet foods (see "Energy Intake: Human Studies"). In addition, none of the animal studies cited here examined food intake after cessation of nicotine. One cannot simply assume that if nicotine decreases general or specific food intake, cessation of nicotine will have the exact opposite effect. With the limitations of previous animal studies in mind, Grunberg (1982) examined the effects of nicotine on body weight and consummatory behavior of rats. Unlike the previous animal experiments, subjects in this study had continuous access to laboratory chow, water, and three different sugar solutions (10%, 25%, and 35% w/v d-glucose). Because nicotine injections might cause decreased food intake by making subjects sick or uncomfortable, Grunberg (1982) administered nicotine via Alzet osmotic minipumps that were implanted subcutaneously between the withers (shoulder blades) of the rats. These devices provide a slow, constant release (roughly 0.5 µUhr) of their contents for weeks (Thceuwes and Yum, 1977). Body weight and food consumption were nieasured daily for 2-week periods before, during, and after drug administration. Animals received
Page 23: rcy63d00
294 NEIL E. GRUNBERG either 0 (i.e., physiological saline), 2.5, 5.0, or 10.0 mg nicotine per kg body weight per day during the drug administration period. Nicotine dihydrochloride was dissolved in physiological saline to make the solutions. Dosages (expressed as nicotine base) were determined from previous research and pilot work to produce body weight effects comparable to those reported for human smokers but to avoid making the rats sick. This paradigm was designed to allow daily measures before, during, and after nicotine administration and to avoid the behavioral and biological stress of daily nicotine injec- tions. Also, this paradigm was designed to allow both between-subjects and within- subject analyses. This precaution was meant to avoid the interpretative difficulties of studies that lacked control groups for time, exposure to experimental procedures, and individual differences. Before receiving saline or nicotine, all experimental groups had similar body weights and consumed similar amounts of the sweet and nonsweet foods. During drug administration, there was an inverse dose-effect relationship between nicotine and body weight: Animals receiving nicotine gained significantly less weight than did the control animals. After cessation of nicotine, animals gained weight at greater rates than did controls. All treatment groups consumed similar amounts of the bland laboratory chow before, during, and after nicotine or saline administration. In contrast, nicotine admin- istration significantly decreased sugar solution consumption. Cessation of nicotine was accompanied by significant increases in sugar solution consumption. These effects were greatest for the highest nicotine group. Moreover, the animals receiving nicotine de- creased consumption of the sweetest solutions during drug administration and increased consumption of the water and lowest concentration sugar solution during drug admin- istration. After cessation of nicotine, they did just the opposite. There were no dif- ferences in total liquid consumption among groups. This experiment clearly found that nicotine affected body weight and sweet food consumption (Grunberg, 1982). In this experiment the decreased caloric intake (from sweet foods) was similar across all nicotine groups, whereas the body weight changes were dose-dependent. Therefore, decrease in caloric intake from sweet foods contributes to the effects of nicotine on body weight, but some additional factor is involved. This interpretation is consistent with the previous animal studies but clarifies that nicotine does decrease energy intake from sweet foods (consistent with the human experiments). In contrast, the increased caloric intake from sweet foods after cessation of nicotine was dose-related in a fashion that could fully explain the dose-related body weight increases after cessa- tion. To summarize the findings of Grunberg's (1982) rat experiment: (1) nicotine decreases sweet food consumption; (2) the effect of nicotine on energy intake contrib- utes to decreased body weight during nicotine administration; (3) cessation of nicotine results in increased sweet food consumption; and (4) the increased energy intake after cessation of nicotine accounts for increased body weight after cessation of nicotine. The human and animal experiments reported by Grunberg (1982) suggest that if only nonsweet foods are available: (1) food intake should not change with nicotine administration or cessation; (2) effects of nicotine on body weight should be reduced; (3) effects of cessation of nicotine on body weight should be reduced or eliminated. Grun- berg et al. (1984) used the animal paradigm of Grunbcrg (1982) to empirically examine the effects of nicotine on body weight and food intake when only bland laboratory chow and water were available (i.e., no sweet foods). Animals were monitored for roughly 2-
Page 24: rcy63d00
INVFRSE RELATIONSHIP BETWEEN TOBACCO USE AND BODY WEIGHT 295 week periods before, during, and after administration of saline or nicotine (4, 8, or 12 mg/kg per day). As predicted, there were no effects of nicotine or cessation of nicotine on food consumption when no sweet foods were available. Body weights were similar before drug administration. Nicotine administration was inversely related to body weight in a dose-response fashion, but the effect of nicotine on body weight was less than when sweet foods were available and food intake changed. After cessation of nicotine, all groups of animals gained weight at similar rates. That is, when no sweet foods were available, there was no effect of nicotine cessation on body weight or food intake. The lack of excessive weight gain after cessation of nicotine when bland food was available was in marked contrast to the increased weight gain following cessation of nicotine when sweet food was available. Grunberg et al. (1984) performed a second experiment that administered nicotine for 3 weeks while only bland food was available. The results were virtually identical to the first experiment. With only bland food available, there were no differences among groups in food intake, body weight was inversely related to nicotine in a dose-effect relationship, and there was no effect of cessation of nicotine on body weight. Because the effects of nicotine on body weight were attenuated when type of food was restricted, changes in consumption of sweet foods must contribute to decreased body weight during nicotine administration. However, because nicotine significantly affected body weight without affecting bland food consumption, nicotine must also affect energy expenditure. Because nicotine cessation did not cause body weight gains when bland food was available but did cause increased body weight and food consump- tion when sweet foods were available, changes in energy intake are critical to excessive weight gains following cessation of nicotine (Grunberg, 1982; Grunberg et al., 1984). The sweet foods in Grunberg's (1982) human and animal experiments covaried sweet taste and carbohydrate content. To determine whether nicotine affects consump- tion of sweet-tasting foods regardless of nutrient or caloric content, and to determine whether nicotine affects consumption of carbohydrates that are not sweet-tasting, Grun- berg et al. (1985) conducted two rat experiments. All animals had continuous access to two foods. In both experiments one food was a low-calorie, nonsweet laboratory chow. In one experiment, the second food was a low-calorie, saccharin-sweetened laboratory chow. In the other experiment, the second food was a nonsweet laboratory chow with carbohydrate (maltose dextrin) added. Body weight and food intake were measured daily for roughly 2-week periods before, during, and after saline or nicotine (6 or 12 mg/kg per day) administration. In both experiments, there was an inverse dose-effect relationship between nic- otine and body weight. After cessation of nicotine, animals gained weight at greater rates than did control animals. Nicotine administration decreased consumption of the sweet, low-carbohydrate food and cessation of nicotine markedly increased consump- tion of this food compared to controls. It is clear that nicotine administration and cessation alters sweet food consumption; sweet taste is important. Nicotine administra- tion and cessation had smaller, but similar effects on carbohydrate consumption. There- fore, carbohydrate content also is involved in nicotine's effects on energy intake. Not surprisingly, the effects of nicotine on body weight were less in the experiment that provided sweet, low-calorie foods. This result further supports the previous conclusion that changes in energy intake contribute to the effects of nicotine on body weight.
Page 25: rcy63d00
296 NEIL E. GRUNBEkG However, as indicated earlier, administration of nicotine and tobacco smoke also can decrease body weight when food intake does not change. This fact has been reported in the 1980s in other laboratories in addition to that of Grunberg and co-workers (e.g., Wellman et al., 1986; Wager-Srdar et al., 1984). Therefore, administration of nicotine must somehow alter energy expenditure. Grunberg et al. (1986) pointed out that virtually all of the previous animal studies of nicotine, energy intake, and body weight used male subjects. Women are more likely than men to report concerns about body weight and to report that they smoke to control body weight (Feldman et al., 1985; Page, 1983; Klesges and Klesges, 1988). This difference between men and women may reflect sociocultural pressures and biases that create different concerns about body weight. Alternatively, there may be a sex dif- ference in the effects of nicotine on body weight. To test this possibility, Grunberg et al. (1986) examined the effects of nicotine or saline on body weight and consummatory behavior of female rats that had access only to bland laboratory chow and water. This food was provided because the studies of male rats using this paradigm found no changes in bland food intake, body weight changes during nicotine administration, and no effects on body weight gain after cessation of nicotine (Grunberg et al., 1984). Therefore, any sex difference should become obvious in comparison to these results. In female rats, nicotine administration was inversely related to food intake and body weight, and cessation of nicotine was positively related to food intake and body weight. These pronounced changes were statistically significant and were greater than the effects of nicotine on body weight and food consumption by male rats (Grunberg et al., 1986). Grunberg et al. (1987) replicated and extended these findings by measuring body weight for 4 months after cessation of nicotine. Again, the effects of nicotine on body weight and food intake were greater in female rats than in male rats. The greater body weight gains by female rats after cessation of nicotine persisted throughout the entire experi- ment (i.e., 4 months after cessation). Levin et al. (1987) reported similar effects of nicotine on body weight and food intake of female rats. If these sex differences observed in animals generalize to humans, then the greater concerns reported by many women about weight gains after smoking cessation may reflect a biological difference in nic- otine's effects. This possibility requires empirical evaluation. Overall, the studies on energy intake indicate that nicotine administration to ani- mals and smoking by humans decrease energy intake and cessation of nicotine or smoking increases energy intake. These effects are particularly pronounced when the foods are sweets and carbohydrates. The effects of nicotine on food intake seem to be greater in females than in males. Changes in energy intake contribute to the effects of nicotine and tobacco on body weight. However, energy expenditure also must be in- volved in the body weight changes during nicotine administration. Energy Expenditure Energy expenditure involves behavioral and biological factors. Physical activity uses up energy and burns calories. If physical activity increases without changing energy intake, then some weight may be lost. Based on energy utilization equations, substantial increases in physical activity are required to lose weight (Powers, 1980). In addition to changes in physical activity, energy expenditure involves the rate at which the body
Page 26: rcy63d00
INVERSE RELATIONSHIP BETWEEN TOBACCO USE AND BODY WEIGHT 297 utilizes energy (i.e., metabolism). If nicotine administration or cessation changes spon- taneous physical activity or metabolism, then body weight could change. Somc human and animal studies have addressed these possibilities, although not to the extent or in the detail of the energy intake experiments. Energy Expenditure: Physical Activity. Cross-sectional human studies compar- ing physical activity in smokers and nonsmokers report a weak negative relationship with smoking (Blair et al., 1985; Kannas, 1981) or no relationship at all (Gyntelberg and Meyer, 1974; Stamford et al., 1984a,b; Stephens and Pederson, 1983). Prospective human studies have reported no changes in physical activity accompanying smoking abstinence (Hatsukami et al., 1984; Hofstetter et al., 1986; Klesges et al., 1987; Rodin, 1987; Stamford et al., 1986). Animal studies of acute nicotine injection and short-term physical activity changes have reported increases (Bryson et al., 1981; Schlatter and Battig, 1979, 1981) and decreases (Hatchell and Collins, 1980; Rodgers, 1979) in physical activity. These studies did not examine chronic nicotine administration, did not measure body weight changes, did not examine physical activity for more than 45-min periods, and did not examine effects of nicotine cessation. Therefore, it is impossible to extrapolate from any of these studies to the human condition. Two recent animal experiments used the Grunberg (1982) paradigm and measured physical activity for 24-hr periods from the 2- week periods before, during, and after nicotine or saline administration. Grunberg and Bowen (1985) reported that nicotine increased physical activity in male rats, but these increases were small and occurred after body weight decreases had already shown up. After cessation of nicotine, physical activity decreased and preceded the weight gains. In contrast, Bowen et al. (1986) reported that nicotine administration or cessation did not alter physical activity in female rats, although body weight changes clearly occurred. It appears that physical activity does not contribute to the effects of nicotine or snioking to decrease body weight. Similarly, there is no evidence that physical activity contributes to body weight increases after cessation of smoking, but one study indicates that it may contribute to body weight gains in male rats after cessation of nicotine. Too few animal data are available to reach unequivocal conclusions about chronic nicotine, physical activity, and body weight. The human studies suggest strongly that physical activity is not involved in the effects of smoking or abstinence on body weight. It may be that smokers who quit smoking intentionally increase physical activity to help offset weight gains. If this is occurring against a backdrop of naturally decreased activity, then it is likely that no changes will be detected. Based on the available human and animal evidence, there is little reason to presume that physical activity is an important factor in the smoking/body weight relationship. However, intentional increased physical activity may be a useful tool to help reduce unwanted weight gains after cessation of smoking. Energy Expenditure: Metabolism. Human and animal studies have examined effects of nicotine and tobacco smoke on metabolism. Several studies comparing smok- ers and nonsmokers report that, in general, smokers have a higher basal metabolic rate (BMR) and respiratory quotient (RQ) (Schlumm, 1930, 1932; Hadley, 1941-1942; Brozek et al., 1958). Sonie studies report that BMR and RQ increase immediately after smoking (Hicstand et al., 1940; Baroni and Mandrioli, 1952; Haggard and Greenberg, 1934), but other studies report no effects of smoking on these measures (Dill et al., 1934; Larson et al., 1961). More recent studies report that acute nicotine administration
Page 27: rcy63d00
298 NEIL E. GRUNBERG increases metabolism (Ghanem, 1973; Ilebekk et al., 1975; Robinson and York, 1986; Schievelbein et al., 1978; Wennmalm, 1982). None of these studies establish whether these effects of nicotine and tobacco smoke alter body weight in smokers or ex-smokers because most of these studies measured metabolism for short periods of time after smoking, did not measure body weight, and did not monitor either variable after cessa- tion of smoking. A few studies have measured metabolic rate after cessation of smoking. Schlumm (1932) reported that, for many smokers, after cessation of smoking, BMR decreased and weight gain occurred. Unfortunately, the data provided are sketchy and food intake is not reported. Glauser et al. (1970) reported decreases in oxygen consumption 1 month after smoking cessation. Dallosso and James (1984) reported a small drop in metabolic rate 6 weeks after smoking cessation. Hofstetter et al. (1986) found that the total energy expenditure of smokers was higher during I day when they were allowed to smoke versus I day that they were not allowed to smoke; there were no differences in resting metabolic rate. In contrast, Silbert and Friedlander (1931) reported no changes in metabolism after cessation of tobacco use. Burse and co-workers (1975, 1982) reported no changes in resting metabolic rate 3 weeks after smoking cessation. Robinson and York (1986) reported no changes in total energy expenditure 7 days after smoking cessation. Stamford et al. (1986) reported no changes in oxygen consumption 2 days after smoking cessation. Schmitthenner et al. (1957) reported that nicotine increased total body oxygen consumption in dogs. Bizzi et al. (1972) found that nicotine increases lipolysis of adipose tissue in rats. Ilebekk et al. (1975) reported that cigarette smoke increases total body oxygen consumption in dogs. Wellman et al. (1986) reported that nicotine alone did not alter thermogenesis in rats, but that nicotine plus caffeine increased brown adipose tissue thermogenesis markedly. The effects of nicotine and tobacco smoke on metabolism are not clear. Because nicotine can decrease body weight without significantly changing energy intake (see "Energy Intake") and without significantly changing physical activity (see "Energy Expenditure: Physical Activity"), there must be biological changes in energy expendi- ture that contribute to the effects of nicotine on body weight. Yet the human and animal studies of nicotine and metabolism do not provide consistent, robust findings. A recent review of this topic suggested that the small sample sizes of the human studies of smoking and energy expenditure may not allow sufficient statistical power to detect any small, but important changes in resting metabolic rate (USDHHS, 1988). Perhaps nicotine and tobacco increase energy expenditure of sonie sources of energy (e.g., fats, carbohydrates), but not all sources of energy. If so, then studies may report different findings depending on the measures of metabolism used and the macronutrient content of the foods consumed. At this point, it is logical to conclude that energy expenditure is involved in the nicotine/body weight relationship, but it is not clear exactly how. 5. POTENTIAL MECHANISMS UNDERLYING THE EFFECTS OF NICOTINE ON ENERGY INTAKE AND EXPENDITURE Based on the available empirical evidence: (1) smoking acts to keep body weight down; (2) smoking cessation results in increased body weight; (3) nicotine is responsible
Page 28: rcy63d00
INVERSE RELATIONSHIP BETWEEN TOBACCO USE AND BODY WEIGHT 299 for the effects of smoking on body weight; (4) smoking and nicotine administration decrease food intake, especially of sweets and carbohydrates; (5) cessation of smoking or nicotine increases food intake, especially of sweets and carbohydrates; (6) the de- creased food intake during smoking or nicotine administration partially contributes to the decreased body weight; (7) the increased food intake after cessation of smoking or nicotine largely explains the increased body weight; (8) changes in physical activity play a small or no role in the smoking/body weight or nicotine/body weight relationship; (9) increased biological energy expenditure (e.g., metabolism) during smoking and nicotine administration seems to be involved in decreasing body weight; and (10) decreased biological energy expenditure after cessation of smoking or nicotine may contribute to increased body weight. The effects of smoking and nicotine on energy intake have been extensively studied, and this database allows consideration of the next level of analysis, i.e., mechanisms that may underlie changes in energy intake and specific food consump- tion. The energy expenditure literature is not as clear and mechanisms are only begin- ; ning to be explored. Although the scientific world has only recently convinced itself that smoking is inversely related to body weight, the tobacco industry made use of this well-accepted "fact" in advertisements for Lucky Strikes in the 1920s and, implicitly, in advertise- ments for Virginia Slims in the 1970s. Similarly, while scientists bicker about energy intake and expenditure, everyone "knows" that smokers who quit smoking get fat because they eat more cakes and candies and their metabolism slows down. In the 1920s the Lucky Strike people told us to "Reach for a Lucky instead of a sweet." The empirical evidence now supports what our grandparents assumed to be true. When we turn to the next level of analysis, however, commonly held beliefs give way as careful laboratory analyses reveal important distinctions. Potential mechanisms involved in the smoking/body weight relationship are presented next. Potential Mechanisms Underlying Changes in Energy Intake Taste Perception. It is widely held that increased food consumption, especially of sweets, after cessation of smoking results from changes in taste perception. It seems reasonable that if cigarette smoking blunts the ability to taste foods, then food consump- tion might decrease. Conversely, if smoking cessation is accompanied by improved ability to taste foods, then increased food consumption would be a reasonable accom- paniment. Many studies have examined the effects of smoking on taste acuity. Pangborn and Trabue (1973) reviewed studies performed between 1935 and 1970 on smoking and taste sensitivity. Typically, smokers and nonsmokers tasted low-con- centration flavored solutions (e.g., sweet, salty, bitter) that were close to the lower limits of taste detectability: 11 studies reported that heavy smoking impaired taste acuity; five studies reported no differences between groups. Even the studies that re- ported that smokers had impaired taste sensitivity were not consistent across taste classes. For example, Krut et al. (1961) reported that smokers had higher thresholds for bitter, but that smokers and nonsmokers did not differ for sweet, sour, or salt. Kaplan and colleagues (Fischer et al., 1963; Kaplan and Glanville, 1964; Kaplan et al., 1965) also reported that smoking is related to higher thresholds for bitter. Peterson et al. (1968) reported that ability to perceive bitter taste improved in smokers who quit smoking, but the ex-smokers' ability was similar to that of a group of smokers who kept smoking.
Page 29: rcy63d00
300 NEIL E. GRUNB,ERG Pursell et al. (1973) questioned the results of these previous studies because they did not consider the subjects' response criteria. Using two measures of sensitivity based on signal detection theory (area under ROC curve and d), Pursell et al. (1973) found that smokers were significantly more sensitive to near threshold solutions and were somewhat more sensitive to suprathreshold solutions. Based on the percent of correct responses, there were no differences in taste sensitivity between smokers and non- smokers. McBurney and Moskat (1975) also used more careful psychophysical methods (including up-down-transformed-response method) to study taste thresholds in smok- ers and nonsmokers. Smokers were slightly less sensitive to bitter and slightly more sensitive to sweet, but the effects were small, inconsistent, and not significant. It is unlikely that the small or nonexistent changes in taste threshold accompanying smoking would result in significant changes in sweet food intake. The smoking and taste threshold studies indicate that smoking may slightly reduce ability to taste bitter foods, but that's about it for smoking and taste sensitivity. Adapta- tion to bitter potentiates the taste of sweet foods and makes water taste sweet (McBurney and Moskat, 1975). Therefore, the effects of smoking on ability to taste bitter may affect the perception of sweet foods, but the effect is not documented and would be minimal or nonexistent. Further, there is no evidence that nicotine affects bitter or any other taste sensitivity, yet animal studies with nicotine similarly show altered preference for sweet foods. The effects of smoking and nicotine on consumption of sweet foods cannot be explained by altered taste perception. This conclusion is consistent with the generaliza- tion that taste sensitivity accounts for little variance in food selection (Rozin, 1984). Intentional Food Restriction or Choice. General and specific food consumption change when foods are intentionally sought or rejected to purposely manipulate a healthful, restricted, or philosophically based diet. Such intentional changes in food consumption are unlikely to explain the inverse relationship between nicotine and sweet food consumption. It seems reasonable to conclude that laboratory rats did not inten- tionally change specific food consumption to control their weight or diet. Moreover, if intentional manipulation of diet explained the human results, one would expect people to purposely avoid sweet, high-caloric foods after cessation of smoking to avoid weight gains. They certainly would not purposely seek fattening foods as their weight crept up. If anything, restraint in consumption of high-caloric foods should act (and may have in the longitudinal studies) to attenuate the increased sweet food consumption after cessa- tion of smoking. Food Preferences. A third possible explanation for changes in energy intake is that preference for and enjoyment of specific foods change with nicotine administration and cessation. As discussed earlier, Grunberg (1982) found that human smokers allowed to smoke ate less sweet food and were less likely to choose sweet foods to eat later in the experimental session. Smokers not allowed to smoke ate more sweets and, compared to smokers smoking and nonsmokers, were niost likely to choose sweet foods to eat again. Further, Grunberg (1980, 1982) found that all three experimental groups rated the foods similarly on taste perception scales. The smokers and nonsmokers judged the intensity of food tastes similarly, but expressed significant differences in food preferences that paralleled the actual food intake. Unlike the taste threshold studies discussed earlier, Grunberg (1982) used foods rather than near-threshold solutions. As a result, these results may reflect a combination of suprathreshold tastes.
Page 30: rcy63d00
•INVERSE RELATIONSHIP BETWEEN TOBACCO USE AND BODY WEIGHT 301 To further compare the effects of smoking on taste perception versus taste prefer- cnces, as distinguished by Grunberg (1982), Redington (1984) had smokers and non- smokers rate intensity and pleasantness of near-threshold solutions before and after a glucose load. Smokers reported decreased hedonic judgment of sweet solutions after a glucose load; the intensity ratings were similar for all subjects. More recently, Rodin (1987) tested the pleasantness/intensity distinction in a prospective study of smoking cessation. Again, preference for sweets increased after smoking cessation, but judg- ments of taste intensity did not change. Based on these studies, it seems that smoking decreases the pleasantness of sweet foods and that smoking cessation increases this variable. The increased sweet preference after cessation of smoking overshoots the preference expressed by nonsmokers (Grun- berg, 1982). These changes in preference for specific tastes parallel the changes in specific food intake. The fact that animals during nicotine administration and cessation show similar changes in sweet food consumption suggests that nicotine acts to alter preference for sweets which translates into changed consumption, unless an individual (e.g., an ex-smoker trying to avoid weight gains) purposely self-restrains to control sweet food consumption. This discussion of nicotine and food preferences can be cast in the broader context of food selection or specific appetites. Differences in food selection involve biological and cultural factors (Rozin, 1984). In the case of nicotine and specific food consump- tion, the rapidity of the changed preferences revealed in human and animal laboratory studies strongly suggests that biological factors underlie these behavioral changes. Learning and expectations certainly may be involved in changes in sweet food prefer- ences with snioking and after cessation. However, the increased consumption of sweet foods by smokers hours after cessation and the marked changes in sweet food consump- tion by animals soon after nicotine administration or cessation strongly suggest a biolog- ical basis for these effects. To understand why nicotine affects specific food preferences requires a digression to the food selection literature. The careful animal experiments by Curt Richter (1939, 1943) clearly established that changes in nutrient and electrolyte levels in the body affect food selection. The notion that specific appetites are a mechanism by which the body maintains intake of necessary foodstuffs and makes adjustments based on needs grew largely out of Richter's work and a study by Clara Davis (1928) of human infants and food selection. Basically, the idea is that humans and animals consume the nutrients they need to maintain a functioning body. Specific appetites develop in response to these needs. Because certain tastes are associated with intake of particular needs (e.g., salt for sodium), tastes are reinforced after consumption of the necessary foodstuff and the organism learns to choose that foodstuff. P. T. Young went beyond stimulus-response (S-R) theory in interpreting Richter's findings about biologically based drives for specific food selection. According to Young (1955), biological need results in an altered hedonic or affective tone such that a specific food may be craved or taste particularly good. Therefore, a specific appetite results from a biological need giving rise to altered hedonic response to specific tastes or foodstuffs, which, in turn, translates into altered food choice and consumption. Young's (1936) "hedonic theory" was the basis for this reinterpretation of Richter's work. At first glance, the differences in interpretation may seem subtle, but they are not. At a time
Page 31: rcy63d00
302 NEIL E. GRUNBERG when S-R theory reigned supreme, Young (1936) applied "hedonic theory" to the results of animal as well as human experiments. Young's (1955) argument for an affective component to specific appetites is convincing as he points out the limitations of a strict S-R interpretation of motivation. In the present context, Young's point again becomes relevant because the time course of altered food preferences with nicotine administration and cessation is short or immediate and humans report clear changes in preferences for which foods taste better. Also relevant to understand nicotine's effects on specific food preferences is Cabanac's (1971, 1979) notion of "alliesthesia" or "changed sensation." Basically, this term refers to the phenomenon in which a particular food is perceived as pleasant or unpleasant depending on the internal state ("milieu interieur") of the body. Cabanac's (1971) "alliesthesia" is similar to Young's (1955) reinterpretation of Richter's (1943) experiments. However, whereas Richter studied rats and (mostly) salt selection, Cabanac studied humans and sweet selection. For example, after ingestion of glucose, humans report that formerly pleasant sweet solutions become unpleasant (Cabanac, 1971). The work of Richter, Young, and Cabanac may apply to the case of nicotine and specific food preferences. Following from Richter and Young, nicotine may act to alter levels or availability of specific nutrients that translate into altered preferences for sweet carbohydrates. Following from Cabanac, the change in preferences for sweet taste may involve a change in glucose availability. Consistent with this possibility, Larson et al. (1961) noted that 27 of 30 animal studies reviewed reported transient, but significant, increases in blood sugar after nicotine injection. However, with respect to humans and smoking, Larson et al. (1961) wrote: "In contrast to the near unanimity of results following the injection of controlled doses of nicotine into animals, the effect of smok- ing on the blood-sugar level in man is notably inconsistent" (p. 325). Based on these studies, effects of nicotine on glucose may or may not underlie changes in preference for sweet foods. A related possibility is that nicotine alters insulin levels in the body and thereby alters preference for sweet foods (possibly by altering glucose transport). Jacobs (1958) reported that insulin injections to rats resulted in increased preference for sweet foods. Briese and Quijada (1979) reported similar results in humans. Kanarek et al. (1980) reported that insulin injections to rats increased carbohydrate consumption. With these studies and theories as background, Grunberg et al. (1988b) examined the effects of nicotine administration on blood levels of glucose and insulin in rats. Unlike the previous animal studies of nicotine and glucose that examined the effects of acute nicotine injections, Grunberg et al. (1988b) administered nicotine or saline contin- uously for 2 weeks via osmotic minipump. As in previous studies, nicotine administra- tion was accompanied by a short-lived increase in blood glucose. This rise in blood glucose disappeared within a day or so before body weight was affected by nicotine. In contrast, nicotine decreased blood levels of insulin, and this decreased level persisted throughout the experiment. Grunberg et al. (1988b) suggested that these changes in blood insulin may underlie the effects of nicotine on sweet food preferences. Nicotine also changes consumption of carbohydrates even when they are not sweet- tasting (Grunberg et al., 1985). Again, if Richter's "we eat what we need" and Young's "what we need tastes good" principles apply, then there must be an effect of
Page 32: rcy63d00
INVERSE RELATIONSHIP BETWEEN TOBACCO USE AND BODY WEIGHT 303 nicotine that alters need for carbohydrates. Because carbohydrates break down in the body by glycolysis to provide glucose and in light of Kanarek et al.'s (1980) finding that insulin injections increase carbohydrate intake, Grunberg et al.'s (1988b) findings for nicotine and insulin also may underlie carbohydrate craving. However, there is another potential mechanism. It has been suggested that increased carbohydrate consumption results in increased levels of the neurotransmitter serotonin in the body (Fernstrom and Wurtman, 1971a,b, 1972, 1973; Wurtman, 1982; Wurtman et al., 1981). In animals, nicotine administration reduces serotonin turnover, resulting in increased central levels (Balfour et al., 1975; Fuxe et al., 1987). Therefore, it may be that nicotine's effects on serotonin result in altered preference for carbohydrates. This idea is consistent with reports that drugs that enhance serotonin transmission decrease carbohydrate consump- tion (Wurtman and Wurtman, 1979). Interestingly, there seem to be effects of insulin on serotonin availability (Wurtman, 1987). Therefore, it is reasonable to postulate that both insulin and serotonin are involved in the effects of nicotine on specific food preferences and that these biochemicals may act via a common mechanism. If this postulate is correct, then Richter's notion of selective appetites now can be extended to the level of food selection to modulate levels of specific neurotransmitters. P. T. Young should not be forgotten in this context either, although he is rarely c ited in recent discussions of foods and mood. Based on Fernstrom, R. Wurtman, and J. Wurtman's work on carbohydrates and serotonin, there have been human studies of the effects of food on mood and performance (Spring, 1984; Spring et al., 1982/1983, 1984). Basically, these studies report that carbohydrate consumption is followed by feelings of calm or sleepiness (Spring et al., 1984). Grunberg (1986b) applied these findings to nicotine and specific appetites by suggesting that cessation of nicotine (or smoking) is accompanied by increased carbohydrate consumption to alleviate the un- pleasantness induced by nicotine abstinence. In effect, carbohydrate consumption might be viewed as a form of self-medication to offset the undesired pharmacological actions of nicotine abstinence. Based on Wurtman's animal studies, Spring's human studies, and Grunberg's (1986b) reasoning for smoking following from the work of Fernstrom, R. Wurtman, J. Wurtman, and Spring, Bowen et al. (D. J. Bowen, B. Spring, and E. Fox, personal communication) assigned 31 smokers who agreed to quit smoking to two daily diet groups: The experimental group consumed a high-carbohydrate, low-protein diet (7/ 1 ratio of carbohydrate/protein) plus tryptophan (hypothetically to raise serotonin and attenuate the unpleasantness of withdrawal); the control group consumed a placebo and a]ow-carbohydrate diet (1/1 carbohydrate/protein ratio). Two weeks after the quit- smoking date, 75% of the high-carbohydrate group were abstinent compared to 47% of the control group. In addition, the high-carbohydrate group reported less anxiety on the Multiple Affect Adjective Checklist (Zuckerman and Lubin, 1965) and less severe withdrawal symptoms as measured by the Smokers Complaint Scale (Schneider and Jarvik, 1984). To summarize potential mechanisms underlying energy intake, there is evidence that changes in insulin levels may be involved in the changes in specific food prefer- ences for sweets and carbohydrates. In addition, altered preference for carbohydrates may involve effects of these foodstuffs on mood and on serotonin levels. These pos- sibilities are consistent with studies of selective appetites and nutritive needs. Effects of nicotine on other neurotransmitters and biochemicals (e.g., dopamine, norepinephrine,
Page 33: rcy63d00
304 NEIL E. GRUNBERG neuropeptide Y) also may be involved in the effects of nicotine on energy intake, but these possibilities have not yet been explored. Potential Mechanisms Underlying Changes in Energy Expenditure In contrast to the extensive consideration of mechanisms underlying nicotine's actions on food intake, there has been minimal attention to potential mechanisms under- lying nicotine and energy expenditure. The existing evidence points to an important role for energy expenditure in the nicotine/body weight relationship, but few studies or investigations have gone beyond the "what happens" stage and on to the "why it happens stage." The Grunberg et al. (19886) study described earlier was designed with this limitation in mind. Insulin and catecholamines were measured daily during 2 weeks of nicotine administration to rats. In addition to the postulate concerning insulin and taste preferences, these biochemicals were measured because they affect energy utiliza- tion. Nicotine administration was accompanied by increased blood levels of cate- cholamines, but this increase was greatly attenuated after a week. Increased cate- cholamines increase energy utilization, so these changes may have contributed to the effects of nicotine on body weight. However, the fact that these biochemical changes did not last makes it unlikely that these effects play an important role in nicotine's effects on body weight. Nicotine administration was also accompanied by decreased blood levels of insulin that persisted throughout the experiment. These changes are consistent with a decrease in lipogenesis. Possibly, the effects of nicotine on insulin are involved in changes in energy expenditure as well as changes in energy intake. This possibility deserves further research attention. Another mechanism that may underlie effects of nicotine on energy expenditure involves enzymatic activity. Carney and Goldberg (1984) reported a significant, positive correlation between body weight gains in ex-smokers and precessation lipoprotein lipase (1.PL) activity in adipose tissue. This finding has generated interest in this enzyme and its role in the smoking/body weight phenomenon. Experiments need to be performed that indicate more than a correlational relationship for LPL and smoking-related body weight changes. Enzymatic changes may be an important mechanism to consider. 6. COMMONALITIES WITH OTHER ADDICTIVE DRUGS If there presently is a zeitgcisl regarding drugs of addiction, then it is that there are common mechanisms involved in the actions and effects of these drugs. That is not to say, however, that all addictive agents act identically. But there is value in comparing the effects of these drugs on behavior and in exploring mechanisms that may account for some of the actions of these drugs. In this context, it is intriguing that the effects of nicotine on food intake, and on sweet carbohydrate intake in particular, are not unique to nicotine. For example, similar to the effects of nicotine, oral self-administration of morphine by rats is accompanied by decreased carbohydrate consumption; cessation of morphine results in increased carbohydrate consumption (Marks-Kaufman and Lipeles, 1982). Morphine injections to rats also decrease carbohydrate intake while increasing fat intake (Marks-Kaufman, 1982). Conversely, when sugar is consumed, rats addicted to
Page 34: rcy63d00
INVERSE RELATIONSHIP BETWEEN TOBACCO USE AND BODY WEIGHT 305 morphine consume less morphine; when sugar is removed, morphine intake increases (Kanarek and Marks-Kaufman, 1988). This finding is consistent with the fact that increased food consumption decreases self-administration of psychoactive drugs and that decreased food consumption increases self-administration of these drugs (Carroll and Meisch, 1984). It also has been reported that there is an inverse relationship between sucrose and alcohol consumption in rats (Lester and Greenberg, 1952; Samson et al., 1982). Like Bowen et al.'s (personal communication) findings with diet and smoking abstinence, Yung et al. (1983) reported that alcoholics who remained sober for 30 days consumed significantly more sugar and carbohydrates than did individuals who relapsed to alcohol use in this period. Perhaps specific foods should be used in treatment programs to help avoid relapse to addictive drugs. Comparisons of drugs of addiction and eating behavior may reveal commonalities underlying substances of abuse. Grunberg and Baum (1985) suggested that drugs that are addicting may "short-circuit" neurochemical pathways that underlie control of body weight and food intake. With repeated administration of these drugs, the body may adapt such that the drugs are "misinterpreted" as necessary foodstuffs. Then, the mechanisms of food selection and regulation may conie to govern drug self-administra- tion as well. This possibility is consistent with the available literature but clearly re- yuires extensive empirical examination. Kanarek and Marks-Kaufman (1988) similarly point out that the relationship between food intake and drug-seeking behavior reveals important similarities that may help to elucidate biological bases of addiction and food selection. Future research that addresses food selection and interaction with pharmacological agents must be carefully designed so that interpretation of results is clear. As discussed by Blundell (1983) in a comprehensive review of food selection studies, many meth- odological concerns must be considered in designing and conducting such studies. In any case, the potential contributions of studies of drugs and body weight to better understand addictive behaviors is vast and deserves substantial research attention. It is also important to remember that people who use addictive drugs commonly use more than one. In the present context, it is relevant that tobacco and alcohol use are highly correlated and that both of these drugs can affect body weight (Kozlowski et al., 1986; Schoenborn and Benson, 1988; Williamson et al., 1987). Therefore, cross-sec- tional and longitudinal analyses of the effects of tobacco use on body weight, energy intake, and energy expenditure should carefully consider and partial out effects of alcohol and other drugs on these variables. Moreover, laboratory experiments should examine whether effects of these drugs are additive, interactive, or interchangeable. Such studies might provide a better understanding of similarities and differences in mechanisms of addictive drug actions. 7. DIGEST Cigarette smoking is inversely related to body weight. This relationship results from effects of nicotine on energy intake and energy expenditure. The relative contribu- tion of changes in energy intake in the nicotine/body weight relationship appears to be
Page 35: rcy63d00
306 NEIL E. GRUNBERG particularly important in accounting for weight gains after cessation of nicotine admin- - istration but also contributes to decreased body weight during nicotine administration. The effects of nicotine on energy intake are most pronounced for sweet foods and carbohydrates. These effects of nicotine on sweet carbohydrates are similar to effects of opiates and alcohol. The effects of these drugs of addiction on specific food selection may reveal biological processes that underlie drug addiction in general and that relate to body weight control. REFERENCES Albanes, D. M.. Jones, Y., Micozzi, N1. S., and Mattson, M. E., 1987, Associations between smoking and body weight in the US population: Analysis of NHANES 11, Am. J. PublicHealdr77 (4):439- 444. Anderson, W. G., 1914, Smoking is detrimental, quoted in Med. Times 42:167-173. Andrews, J., and McGarry, J. M., 1972, A community study of smoking in pregnancy, J. Obstet. Gvnaecol. Br. Conmmnw. 79(12):1057-1073. Ashford, J. R., Brown, S., Duffield, D. P., Smith, C. S., and Fay, J. W. J., 1961, The relation between smoking habits and physique, respiratory symptoms, ventilatory function, and radiological pncu- moconiosis amongst coal workers at three Scottish collieries, Br. J. Prev. Soc. Med. 1.5:106-116. Averett, J. W., 1962, The effects of nicotine on weight increment, activity, food intake, and water intake in weanling albino rats, Diss. Abst. lnt. 23:2187. Balfour, I). J. K., Khuller, A. K., and Longden, A., 1975, Effects of nicotine on plasma cortico- sterone and brain amines in stressed and unstressed rats, PhnnnaroL Bioclrem. Behov. 3(2):179- 184. Baroni, 1V., and Mandrioli, C., 1952, F.ffetti del fumo di sigaretta sul metabolismo di base, Policlinico 59:101-103. Becker, R. F., and King, J. E., 1966, Effects of nicotine absorption in rats during pregnancy, Science 154:417-418. Biener, K., 1981, Exraucherinnen [Women who have stopped smoking], Munchetr. Med. Nochetrschr. 123(25):1035-1038. Birch, D., 1975, Control: Cigarettes and calories, Can. Nurse 71(3):33-55. Bizzi, A., Tacconi, M. T., Medea, A., and Garattini. S., 1972, Sonic aspects of the effect of nicotine in plasma FFA and tissue triglycerides, Phannucnlogy 7:216-224. Bjclke, W., 1971, Variation in height and weight in the Norwegian population, Br. J. Prer. Soc. Med. 25:192-202. Blackburn, H., Brozek, J., and Taylor. H. 1.., 1960, Common circulatory measurements in smokers and nonsrnokers, Circulation 22:1112-1124. Blair, A., Blair, S. N., Howe, H. G., Pate, R. R., Rosenberg, M., Parker, G. 1v1., and Pickle, l.. W., 1980, Physical, psychological, and sociodemographic differences among smokers, exsmokers, and nonsnrokers in a working population, Prev. Med. 9(6):747-759. Blair, S. N.. Jacobs, f). R., and Powell. K. F., 1985. Relationships between exercise or physical activity and other health behaviors, Public flealth Rep. 100(2):172-180. Blitzer, 1'. H., Rinmr, A. A., and Giefer, E. E., 1977, The effect of cessation of snroking on body weight in 57.032 women: Cross-sectional and longitudinal analyses. J. Cltrnnic Dis. 30(7):415- 429. Blundcll, J. 1?., 1983, Problems and processes underlying the control of food selection and nutrient intake, in: Nutrition and Brain, Voluroe 6 (R. J. Wurtman and J. J. Wurtman, eds.), pp. 163-221, Raven Press, New York. Bogen, 1?., 1929, The composition of cigarettes and cigarette srnoke. JAMA 93:1 110-1114.
Page 36: rcy63d00
i 1 INVERSE RELATIONSHIP BETWEEN TOBACCO USE AND BODY WEIGHT 307 Bosse. R., Garvey, A. J., and Costa, P. T., Jr., 1980, Predictors of weight change following smoking cessation, lnt. J. Addictions 15(7):969-991. Bowen, D. J.. Eury, S. E., and Grunberg, N. E., 1986, Nicotine's effects on female rats' body weight: Caloric intake and physical activity, Pharnnacol. Biochern. Behav. 25(6):1131-1136. Briese, E., and Quijada, M., 1979, Positive alliesthesia after insulin, Experientia 35:1058-1059. Brozek, J., Blackburn, H., Taylor, H. L., and Keys, A., 1958, Pulse rate and blood pressure in male cigarette sniokers and non-smokers. Delivered at Third World Congress of Cardiology. Bryson, R., Biner, O. M., McNair, E., Bergondy, M., and Abrams, O. R., 1981, Effects of nicotine on two types of motor activity in rats, Psvchnpharmacology 73:168-170. Burse, R. L., Bynuni, G. D., Pandolf, K. B., Goldman, R. F., Simms, E. A. H., and Danforth, E. R., 1975, Increased appetite and unchanged metabolism upon cessation of smoking with diet held constant, Physiologist 18:157. Burse, R. L., Goldman, R. F., Danforth, E., Horton, E. S., and Sims, E. A. H., 1982, Effects of cigarette smoking on body weight, energy expenditure, appetite, and endocrine function, U.S. Amny Medical Research and Development Command, Fort Detrick, Frederick, Maryland, Report #M25/82, NTIS No. AD-A114 213/2. Cabanac, M., 1971, Physiological role of pleasure, Science 173:1103-1107. Cabanac, M., 1979, Sensory pleasure, Q. Rev. Biol. 54:1-29. Cambien, F., Richard, J. L., Ducimetiere, P., Warnet, J. M., and Kahn, J., 1982, The Paris car- diovascular risk factor prevention trial. Effects of two years of intervention in a population of young men, J. Epiderniol. Commun. Health 35(2):91-97. Carney, R. M., and Goldberg, A. P., 1984, Weight gain after cessation of cigarette smoking. A possible role for adipose-tissue lipoprotein lipase, N. F.'ngl. J. Med. 310(10):614-616. Carroll, M. F., and Meisch, R. A., 1984, Increased drug-reinforced behavior due to food deprivation, in: Advances in Behavioral Pharmacology, Volume 4 (T. Thompson, P. Dews, and J. E. Barrett, cds. ), pp. 47-88, Academic Press, New York. Charlton, A., 1984, The Brigantia Smoking Survey: A general review, in: Public Education about Cancer. Recent Research and Current Programmes (P. Hobbs, ed.), pp. 92-102, UICC Technical Report Series-Volume 77. Coates, T. J., and Li, V., 1983, Does smoking cessation lead to weight gain?: The experience of asbestos-exposed shipyard workers, Am. J. Public Health 73:1303-1304. Comstock, G. W., and Stone, R. W., 1972, Changes in body weight and subcutaneous fatness related to smoking habits, Arch. Environ. Health 24(4):271-276. Dallosso, H. M., and James, W. P. T., 1984, The role of smoking in the regulation of energy balance, Int. J. Obesity 8(4):365-375. Davis, C., 1928, Self-selection of diets by newly-weaned infants, Am. J. Dis. Children 36:651-679. I)iehl, H. S., 1929, The physique of sniokers as compared to non-smokers, Minn. Mcd. 12:424-427. Dill, D. B., E:dwards, H. T., and Forbes, W. H., 1934, Tobacco smoking in relation to blood sugar, blood lactic acid and metabolism, Am. J. Physiol, 109:118-122. Emont, S. L., and Cummings, K. M., 1987, Weight gain following smoking cessation: A possible role for nicotine replacement in weight management, Addictive Behav. 12:151-155. Esser, J., 1903, Die Beziehungen des Nervus vagus su F.rkrankungen von Herz und Lungen, speziell bei experimenteller chronischer Nikotinvergiftung, Arch. Exp. Pathol. Berlin 49:190-212. P.vans, J. R., Hughes, R. E., and Jones, P. R., 1967, Sonie effects of cigarette smoke on guinea-pigs, Proc. Nutr. Soc. 26:xxxvi. Fagcrstrom, K. 0., 1987, Reducing the weight gain after stopping smoking, Addictive Behav. 12:91-93. Favarger, H., 1906, Zur frage der chronischen Tabakver-giftung, Wien. Klin. Wochenschr. 19:636-637. Fchily, A. M., Phillips, K. M., and Yamell, W. G., 1984, I)iet, smoking, social class, and body mass index in the Caerphilly Heart Disease Study, Arn. J. Clin. Ntctr. 40(4):827-833. Feldman, W., Hodgson, C., and Corber, S., 1985, Relationship between higher prevalence of smoking and we~;ht concern amongst adolescent girls, Can. J. Public Health 76:205-206. Fenistrom, J. I)., and Wurtman, R. J., 1971a, Brain serotonin content: Physiological dependence on plasma tryptophan levels, Science 173:149-152.
Page 37: rcy63d00
308 NEIL E. GRUNBERG Fernstrom, J. D., and Wurtman, R. J., 1971b, Brain serotonin content: Increase following injection of carbohydrate diet, Science 174:1023-1025. Fernstrom, J. D., and Wurtman, R. J., 1972, Brain serotonin content: Physiological regulation by plasma neutral amino acids, Science 178:414-416. Fernstrom,l. D., and Wurtman, R. J., 1973, Control of brain 5-HT content by dietary carbohydrates, in: Serotonin and Behavior (J. Barchas and E. Usdin, eds.), pp. 121-128, Academic Press, New York. Fink, B., 1921, Smoking and scholarship again, School Soc. 14:87-89. Fischer, R., Griffen, F., and Kaplan, A. R., 1963, Taste thresholds, cigarette smoking, and food dislikes, Med. Exp. 9:151-169. Fisher, M., and Gordon, T., 1985, The relation of drinking and smoking habits to diet: The Lipid Research Clinics Prevalence Study, Am. J. Clin. NtUr. 41(3):623-630. Friedman, G. D., and Siegelaub, A. B., 1980, Changes after quitting cigarette smoking, Circulation 61: 716-723. Friedman, G. D., King, M-C., Klatsky, A. L., and Hulley, S. B., 1981, Characteristics of smoking- discordant monozygotic twins, in: Twin Research 3, Part C: Epidemiological and Clinical Studies (L. Gedda, P. Parisi, and W. E. Nance, eds.), pp. 17-22, Alan R. Liss, New York. Fuxe, K., Andersson, K., Eneroth, P., Harfstrand, A., Nordberg, A., and Agnati, L. F., 1987, Effects of nicotine and exposure to cigarette smoke on discrete dopamine and noradrenaline nerve terniinal systems of the telencephalon and diencephalon of the rat. Relationship to reward mechanisms and distribution of nicotine binding sites in brain, in: Tobacco Smoking and Nicotine. A Netnobiological Approach (W. R. Martin, G. R. Van Loon, E. T. Iwamoto, and L. Davis, eds.), pp. 225-262, Plenum Press, New York. Garn, S. M., Shaw, H. A., and McCabe, K. E., 1978, Effect of maternal smoking on weight and weight gain between pregnancies (Letter), Am. J. Clin. Nutr. 3l(8):1302-1303. Garrison, R. J., Feinleib, M., Castelli, W. P., and McNamara, P. M., 1983, Cigarette smoking as a confounder of the relationship between relative weight and long-term mortality. The Framingham Heart Study, JAMA 249(16):2199-2203. Garvey, A. J., Bosse, R., and Seltzer, C. C„ 1974, Smoking, weight change, and age: A longitudinal analysis, Arch. Environ. Health 28(16):327-329. Gaudet, F. J., and Hugh, W. C., Jr., 1969, Concomitant habit changes associated with changes in smoking habits, Med. Times 97:195-205. Ghanem, M. H., 1973, Metabolic effects to tobacco smoking, Alexandria Med. J. 19(3):224-235. Gilbert, R. M., and Pope, M. A., 1982, Early effects of quitting smoking, Psvrhopharmacologv 78: 121-127. Glauser, S. C., Glauser, E. M., Reidenberg, M. M., Rusy, B. F., and Tallarida, R. J., 1970, Meta- bolic changes associated with the cessation of cigarette smoking, Arch. Ern-iron. Health 20(3): 377-381. Goldbourt, U., and Medalie, J. H., 1977, Characteristics of smokers, nonsmokers, and ex-smokers among 10,000 adult males in Israel, Il. Physiologic, biochemical, and genetic characteristics, Am. J. Epidemiol. 105(1):75-86. Gordon, T., Kannel, W. B., Dawber. T. R., and McGee, D., 1975, Changes associated with quitting cigarette smoking: The Framingham study, Am. Heart J. 90(3):322-328. Gormican, A., Valentine, J., and Satter, E., 1980, Relationships of maternal weight gain, prepregnancy weight, and infant birthweight, J. Am. Diet. Assoc. 77(6):662-667. Grinstead, O. A., 1981, Preventing weight gain following smoking cessation: A comparison of behav- ioral treatment approaches. Doctoral dissertation, University of California, !-os Angeles, Thesis No. 82-06024, University Microfilms International, Ann Arbor, M1. Gritz, F.. R., Carr, C. R., and Marcus, A. C., 1988, Unaided smoking cessation: Great American Smokeout and New Year's Day quitters, J. Psvchosoc. Oncol. 6(3/4):217-235. Grossarth-Maticek, R., Kanazier, D. T., Vetter, H., and Jankovic, M., 1983, Smoking as a risk factor for lung cancer and cardiac infarct as mediated by psychosocial variables, Psvchorher. Psvchosoma. 39(2):94-105. Grunberg, N. E., 1980, The effects of nicotine on food consumption and taste preferences (Doctoral dissertation, Coluntbia University), Diss. Ahstr. Int. 41(2):728-B. tn m ~ v tn J F-~ ~ t9
Page 38: rcy63d00
INVERSE RELATIONSHIP BETWEEN TOBACCO USE AND BODY WEIGHT 309 Grunberg, N. E., 1982, The effects of nicotine and cigarette smoking on food consumption and taste preferences, Addictive Behav. 7:317-331. Grunbcrg, N. E., 1986a, Behavioral and biological factors in the relationship between tobacco use and body weight, in: Advances in Behavioral Medicine, Volume 2 (E. S. Katkin and S. B. Manuck, eds.), pp. 97-129, JAI Press, Greenwich, CT. Grunberg, N.E., 1986b, Nicotine as a psychoactive drug: Appetite regulation, Psychopharmacol. Bull. 22(3):875-881. Grunberg, N. E., and Baum, A., 1985, Biological commonalities of stress and substance abuse, in: Coping and Substance Use (S. Shiffman and T. A. Wills, eds.), pp. 25-62, Academic Press, New York. Grunberg, N. E., and Bowen, D. J., 1985, The role of physical activity in nicotine's effects on body weight, Pharmacol. Biochem. Behav. 23:851-854. Grunberg, N. E., and Morse, D. E., 1984, Cigarette smoking and food consumption in the United States, J. Appl. Soc. Psychol. 14:310-317. Grunberg, N. E., Bowen, D. J., and Morse, D. E., 1984, Effects of nicotine on body weight and food consumption in rats, Psychopharrnacology 83:93-98. Grunberg, N. E., Bowen, D. J., Maycock, V. A., and Nespor, S. M., 1985, The importance of sweet taste and caloric content in the effects of nicotine on specific food consumption, Psychopharmacol- oRy 87(2):198-203. Grunberg, N. E., Bowen, D. J., and Winders, S. E., 1986, Effects of nicotine on body weight and food consumption in female rats, PsYchopharmacology 90(I):101-105. Grunberg, N. E., Winders, S. E., and Popp, K. A., 1987, Sex differences in nicotine's effects on consummatory behavior and body weight in rats, Psychopharmacology 91:221-225. Grunberg, N. E., Popp, K. A., and Winders, S. E., 1988a, Effects of nicotine on body weight in rats with access to "junk" foods, Psychopharmacology 94:536-539. Grunberg, N. E., Popp, K. A., Bowen, D. J.. Nespor, S. M., and Winders, S. E., 1988b, Effects of chronic nicotine administration on insulin, glucose, epinephrine, and norepinephrine, Life Sci. 42: 161-170. Gunn, R. C., and Shapiro, A., 1985, Life stress, weight gain, and resuming smoking after success in a cessation clinic, Psychol. Rep. 57(3, Part 2):1035-1039. Gyntelberg, F., and Meyer, J., 1974, Relationship between blood pressure and physical fitness, smok- ing, and alcohol consumption in Copenhagen males aged 40-59, Acta Med. Scand. 195(5):375- 380. ftadley, H. G., 1941, The effect of tobacco upon growth, longevity and metabolism, Med. Rec.153: 350-352. Hadley, H. G., 1941-1942, The effect of tobacco upon the basal metabolic rate, J. Med. 22:121-122. Haggard, H. W., and Greenberg, L. A., 1934, The effects of cigarette smoking upon the blood sugar, Science 79:165-166. Hall, S. M., Ginsberg, D., and Jones, J. T., 1986, Smoking cessation and weight gain, J. Consult. CNn. PsychoL. 54(3):342-346. Hatchcll, P. C., and Collins, A. C., 1980, The influence of genotype and sex on behavioral sensitivity to nicotine in mice, Psychopharmacology 71:45-49. Hatsukami, D. K., Hughes, J. R., Pickens, R. W., and Svikis, D., 1984, Tobacco withdrawal symp- toms: An experimental analysis, Psychophmnnacology 84(2):231-236. Haworth, J. C., Ellestad-Sayed, J. J., King, J., and Dilling, L. A., 1980, Fetal growth retardation in cigarette-smoking mothers is not due to decreased maternal food intake, Am. J. Obstet. Gynecol. 137(6):719-723. Heath, C. W., 1958, Differences between smokers and non-smokers, Arch. /ntern. Med. 101:377-388. Ilickey, N., and Mulcahy, R., 1973, Effect of cessation of smoking on body weight after myocardial infarction, Am. J. Clin. Nutr. 26(4):385-386. Iliestand, W. A., Ramsey, H. J., and Hale, D. M., 1940, The effects of cigarette smoking on metabolic rate, heart rate, oxygen pulse, and breathing rate, J. Lab. Clin. Med. 25:1013-1017. Higgins, M. W., and Kjelsberg, M., 1967, Characteristics of smokers and nonsmokers in Tecumseh, Michigan, Arn. J. Epidemiol. 86:60-77.
Page 39: rcy63d00
310 NEIL E. GRUNBERG Hjermann, I., Hclgeland, A., Holme, I., Lund-Larsen, P. G., and Leren, P., 1976, The intercorrelation of serum cholesterol, cigarette smoking and body weight, Acta Med. Scand. 200:479-485. Hofstetter, A., Schutz, Y., Jequicr, E., and Wahren, J., 1986, Increased 24-hour energy expenditure in cigarette smokers, N. EngL J. Med. 314(2):79-82. Holcomb, H. S., and Meigs, J. W., 1972, Medical absenteeism among cigarette and cigar and pipe smokers, Arch. Environ. Health 25(4):295-300. Holme, I., Hjermann, I., Helgeland, A., and Leren, P., 1985, The Oslo study: Diet and antismoking advice. Additional results from a 5-year primary preventive trial in middle-aged men, Prev. Med. 14(3):279-292. Holt, W. L., 1921, Educational research and statistics. Apparent effects of smoking among University of Tennessee freshmen, School Soc. 14:136-138. Holt, W. L., 1922-1923, A statistical study of smokers and nonsmokers at the University of Tennessee, J. Am. Stat. Assoc. 18:766-772. Howell, R. W., 1971, Obesity and smoking habits (Letter), Br. Med. J. 4(5787):625. Hughes, J. A., and Hutchinson, D. C. S., 1983, Smoking, lung function, and body weight (Letter), Br. Med. J. 286(6369):977. Hughes, R. E., Jones, P. R., and Nicholas, P., 1970, Some effects of experimentally-produced cigarette smoke on the growth, vitamin C metabolism and organ weights of guinea-pigs, J. Pharm. Phar- macol. 22:823-827. Huston, J. J., and Stenson, K., 1974, The development of obesity in fit young men: A regimental survey, Practitioner 212(1271):700-705. Ilebekk, A., Miller, N. E., and Mjos, O. D., 1975, Effects of nicotine and inhalation of cigarette smoke on total body oxygen consumption in dogs, Scand. J. Clin. Lab. /nves. 35(1):67-72. Jacobs, D. R., and Gottenborg, S., 1981, Smoking and weight: The Minnesota Lipid Research Clinic, Am. J. Public Health 71(4):391-396. Jacobs, H. L., 1958, Studies on sugar preference: 1. The preference for glucose solutions and its modification by injections of insulin, J. Comp. Physiol. Psychol. 51:304-310. Jenkins, C. D., Rosenman, R. H., and Zyzanski, S. J., 1968, Cigarette smoking-Its relationship to coronary heart disease and related risk factors in the westem collaborative group study, Circulation 38:1140-1155. Jenkins, C. D., Zyzanski, S. J., & Rosenman, R. H., 1973, Biological, psychological, and social characteristics of men with different smoking habits, Health Serv. Rep. 88(9):834-843. Kanarek, R. B., and Marks-Kaufman, R., 1988, Animal models of appetitive behavior: Interaction of nutritional factors and drug seeking behavior, in: Control of Appetite: Current Concepts rn Nutrition (M. Winick, ed.), pp. 1-25, Wiley, New York. Kanarek, R. B., Marks-Kaufman, R., and Lipcles, B. J., 1980, Increased carbohydrate intake as a function of insulin administration in rats, Physiol. Behav. 25:779-782. Kannas, L., 1981, The dimensions of health behavior among young men in Finland, lnt. J. Health Educ. 24(3):146-155. Kaplan, A. R., and Glanville, E. V., 1964, Taste thresholds for bitterness and cigarette smoking, Nature 202:1366. Kaplan, A. R., Glanville, E. V., and Fischer, R., 1965, Cummulative effect of age and smoking on taste sensitivity in males and females, J. Gerontol. 20:334-337. Kardasevitch, B. I., and Khais, L. M., 1940, The histological changes in the respiratory apparatus of guinea-pigs under the influence of nicotine inhalations in manufacturies, J. Otol. Rhinol. l.aryngol. 17:237-241. Kerschbaum, A., Bellet, S., and Khorsandian, R., 1965, Elevation of serum cholesterol after admin- istration of nicotine, Am. Heart J. 69:206-210. Khosla, T., and Lowe, C. R., 1971, Obesity and smoking habits, Br. Med. J. 4(5778):I0-13. Kim, S. S., 1936, The influence of continuous subcutaneous injection of nicotine or adrenal gland and anterior lobe of hypophysis. 11. Histological studies of the reaction of silver nitric granules of vitamin C. Chosen /gaku-Kwai Zasshi 26:S1. Kitchen, J. M. W., 1889, On the health-value to man of the so-called divinely beneficent gift, tobacco, Med. Rec. 35:459-460.
Page 40: rcy63d00
INVERSE RELATIONSHIP BETWEEN TOBACCO USE AND BODY WEIGHT 311 Kittcl, F., Rustin, R. M., Dramaix, M., DeBacker, G., and Komitzer, M., 1978, Psycho-socio- biological correlates of moderate overweight in an industrial population, J. Psychosorn. Res. 22(3): 145-158. Klesges, R. C., and Klesges, L. M., 1988, Cigarette smoking as a dietary strategy in a university population, Int. J. Eating Dis. 7:413-419. Klesges, R. C., Meyers, A. W., Hanson, C. L., and Eck, L., 1987, Smoking cessation and weight gain in males and females, Presented at the Association for the Advancement of Behavior Therapy, Boston. ` Kopczynski, J., 1972, Height and weight of adults in Cracow, Epidemiol. Rev. 26(4):452-463. Kozlowski, L. T., Jelinek, L. C., and Pope, M. A.. 1986, Cigarette smoking among alcohol abusers: A continuing and neglected problem, Can. J. Public Health 77:205-207. Kramer, J. F., 1982, A one-year follow-up of participants in a Smoke Stoppers program, Patient Counsel. Health Educ. 4(2):89-94. Krut, L. H., Perrin, M. J., and Bronte-Stewart, B., 1961, Taste perception in smokers and non-smokers, Br. Med. J. 1:379-384. l.arson, P. S., Haag, H. B., and Silvette, H., 1961, Tobacco: Experimental and Clinical Studies (2 parts), Williams & Wilkins, Baltimore. Leschtschinskaja, 0., 1926, The toxicology of nicotine (In Russian with German Summary), Bezopas- nosti No. 7-8:26-36. Lester, D., and Greenberg, L. A., 1952, Nutrition and the etiology of alcoholism. The effect of sucrose, saccharin, and fat on the self-selection of ethyl alcohol by rats, Q. J. Stud. Alcohol 13: 553. l.evin, f:. 1)., Morgan, M. M., Galvez. C., and Ellison, G. D., 1987, Chronic nicotine and withdrawal effects on body weight and food and water consumption in female rats, Physiol. Behav. 39:44I- 444. Lickint, F., 1933, Einfluss des Tabaks auf die korperliche und geistige Entwicklung der Schuljugend, Gesundheit Erzieh 46:465-478. Lincoln, J., 1969, Weight gain after cessation of smoking, JAMA 210:1765. I.incoln, J. F.., 1970, Relation of income to body weight in cigarette smokers and nonsmokers (Letter), JAMA 214(6):1121. Lund-Larsen, P. G., and Tretli, S., 1982, Changes in smoking habits and body weight after a three-year- period-The cardiovascular disease study in Finnmark, J. Chronic Dis. 35:773-780. Manley, R. S., and Boland, F. J., 1983, Side-effects and weight gain following a smoking cessation program, Addictive Behav. 8(4):375-380. Marks-Kaufman, R., 1982, Increased fat consumption induced by morphine administration in rats, PharmacoL. Biochem. Behav. 16:949-955. Marks-Kaufman, R., and Lipeles, B. J., 1982, Patterns of nutrient selection in rats orally self-admin- istering morphine, Natr. Behav. 1:33-46. Mertsuya, T., 1982, Alcohol intake and smoking habits of Japanese telephone employees and their effects on the dietary intakes, J. Formosan Med. Assoc. 81(7):857-867. McArthur, C., Waldron, E., and Dickinson, J., 1958, The psychology of smoking, J. Abnorm. Soc. Psvchol. 56:267-275. \1cRurney, I). I1., and Moskat, L. J., 1975, Taste thresholds in college-age smokers and non-smokers, Perception Ps)chop{p•sics 18(2):71-73. McNair, E., and Bryson, R., 1983, Effects of nicotine on weight changes and food consumption in rats, PharmacoL Biochent. Behav. 18(3):341-344. Morgan, M. M., and Ellison, G., 1987, 1)ifferent effects of chronic nicotine treatment regimens on body wcight and tolerance in the rat, Ps)•chopharntacology 91(2):236-238. Munster, G., and Battig, K., 1975, Nicotine-induced hypophagia and hypodipsia in deprived and in hypothalamically stimulated rats, Ps' vchopharntacologia 41:21 I-217. Nemery, B., Moavero, N. E., Brasseur, L., and Stanescu, D. C., 1983, Smoking, lung function, and , body weight, Br. Med. J. 286(6362):249-251. NIH Consensus Development Conference on Health Implications of Smokeless Tobacco Use, Jan. 13- 15, 1986, National Institutes of Health, Bethesda, MD. ~.:YN;tL, _T_~i~lii1 t%N~; OcrVn
Page 41: rcy63d00
312 NEIL E. GRUNBERG Noppa, H., and Bengtsson, C., 1980, Obesity in relation to smoking: A population study of women in Goteborg, Sweden, Prev. dfed. 9(4):534-543. Otis, E. 0., 1884, The use of tobacco by boys, Boston Med. Surg. J. 110:99-103. Page, R. M., 1983, Assessing gender differences in college cigarette smoking intenders and noninten- ders, Doctoral dissertation. Southern Illinois University, Thesis No. 82-21953, University Micro- films International, Ann Arbor,°Ml. Pangborn, R. M., and Trabue, 1. M., 1973, Gustatory responses during periods of controlled and ad lib cigarette smoking, Perception Ps}chophysics 1:139-144. Passey, R. D., Elson, L. A., and Lewis, P. A., 1959, Tobacco smoke and growth rate, Br. Empire Cancer Campaign, Ann. Rep.17:89. Passey, R. D., Elson, L. A., and Connors, T. A., 1961, Growth and metabolic effects of cigarcttc smoking and nicotine inhalation, Br. Empire Cancer Campaign, Ann. Rep. 39(Part II):90. Pemberton, J., and Macleod, K. 1. E., 1956, Rural health survey of men over forty, Public Health Rep. 71:1213-1220. Perlick, D., 1977, The withdrawal syndrome: Nicotine addiction and the effects of stopping smoking in heavy and light smokers. Unpublished doctoral dissertation, Columbia University, New York, Doctoral dissertation, Columbia University, Thesis No. 77-14, 833, University Microfilms Interna- tional, Ann Arbor, MI. Perrin, M. J., Krut, l.. H., and Bronte-Stewart, B., 1961, Smoking and food preferences. Br. Med. J. 1: 387-388. Peters, J. M., and Ferris, B. G., Jr., 1967, Morphological constitution and smoking, Arch. Environ. Health 14:678-681. Peterson, D. L., Longergan, L. H., and Hardinge. M. G., 1968, Smoking and taste perception, Arch. Environ. Health 16:219. Picone, T. A., Allen, L. H., Schramm, M. M., and Olsen, P. N., 1982, Pregnancy outcome in North American women. 1. Effects of diet, cigarette smoking, and psychological stress on maternal weight gain, Arn. J. Clin. Nurr. 36(6):1205-1213. Pincherle, G., 1971, Obesity and smoking habits (Letter), Br. Med. J. 4(5782):298. Powell, D. R., and McCann, B. S., 1981, The effects of a multiple treatment program and maintenance procedures on smoking cessation, Prev. Med. 10:94-104. Powers, P. S., 1980, Obesity, the Regulation of Weight, Williams & Wilkins, Baltimore. Puddey, 1. B., Vandongen, R., Beilin, L. J., English, D. R., and Ukich, A. W., 1985, The effect of stopping smoking on blood pressure-A controlled trial, J. Chronic Dis. 38(6):483-493. Pursell, E. D., Sanders, R. E., and Haude, R. H., 1973, Sensitivity to sucrose in smokers and non- smokers: A comparison of TSD and percent correct measures, Ps' vchopfrysics 14(1):34-36. Rabkin, S., 1984, Relationship between weight change and the reduction or cessation of cigarette smoking, Int. J. Obesity 8(6):665-673. Rantakallio, P., and Hartikainen-Sorri, A., 1981, The relationship between birth weight, smoking during pregnancy and maternal weight gain, Am. J. f:pidemiol. 113(5):590-595. Redington, K., 1984, Taste differences between cigarette smokers and nonsmokers, Pharrnacal. Bio- chenn. Behav. 21:203-208. Richter, C. P., 1939, Salt taste thresholds of normal and adrenalectomized rats, Endocrinology 24:367- 371. Richter, C. P., 1943, Total self regulatory functions in animals and human beings, HarveyLectare Series 38:63-103. Robinson, S., and York, I). A., 1986. The effect of cigarette smoking on the thermic response to feedings, Int. J. Obesitv 10:407-417. Rodgers, R. J., 1979, Effects of nicotine, mecamylamine and hexamethonium on shock-induced fight- ing, pain reactivity and locomotor behavior in rats, PsvchnpharnmcolaRy 66:93-98. Rodin, J., 1987, Weight changes following smoking cessation: The role of food intake and exercise, Addictive Behav. 12:303-317. Rozin, P., 1984, The acquisition of food habits and preferences, in: Behavioral Health: A Handbook of Health Enhancement and 1)isease Prevention (J. D. Matarazzo, S. M. Weiss, J. A. Herd, N. E. Miller, and S. M. Weiss, eds.), pp. 590-607, Wiley, New York.
Page 42: rcy63d00
INVEF;SE RELATIONSHIP BETWEEN TOBACCO USE AND BODY WEIGHT 313 Rush, D.. 1974, Examination of the relationship between birthweight, cigarette smoking during pregnan- cy and maternal weight gain, J. Obstet. Gynaecology Br. Commonw. 81(10):746-752. Samson, H. H., Roehrs, T. A., and Tolliver, G. A., 1982, Ethanol reinforced responding in the rat: A concurrent analysis using sucrose as the alternate choice, Pharmacol. Biochem. Behav. 17:333- 339. Schechter, M. I)., and Cook, P. G., 1976, Nicotine-induced weight loss in rats without an effect on appetite, Eur. J. Pharmacol. 38( l):63-69. Schievelbein, H., Heinemann, G., Loeschenkohl, K., Troll, C., and Schlegel, J., 1978, Metabolic aspects of smoking behavior, in: Smoking Behaviour. Physiological and Psychological /nfluences (R. E. Thornton, ed.). Churchill Livingstone, Edinburgh. Schlatter. J., and Battig, K., 1979, Differential effects of nicotine and amphetamine on locomotor activity and maze exploration in two rat lines, Psychopharmacology 64:155-161. Schlatter, J., and Battig, K., 1981, The adrenergic role in the manifestation of nicotine effects on maze ambulation in Roman High- and Roman Low-Avoidance rats, Br. J. Addict. 76:199-209. Schlumm, F., 1930, Nikotinabusus und grundumsatz, Verh. Deut. Ges. inn. Med. 42:151-155. Schlumm, F., 1932. Tabakmissbrauch und hyperthyreoidismus, Ztschr. Klin. Med. 120:648-653. Schmitthenner, J. E., Hafkenshiel, J. B., Forte, I., and Riegel, C., 1957, The effects of nicotine on left- ventricular work and coronary blood flow, Am. J. Med. Sci. 234:487-488. Schneider, N. G., and Jarvik, M. E., 1984, Time course of smoking withdrawal symptoms as a function of nicotine replacement, Psychopharmacology 82:143-144. Schoenbom, C. A., and Benson, V., 1988, Relationships between smoking and other unhealthy habits: United States, 1985, NCHS Advancedata 154:1-8. Schoenenberger, J. C., 1982, Smoking change in relation to changes in blood pressure, weight, and cholesterol, Prev. Med. 11(4):441-453. Seaver, J. W., 1897, The effects of nicotine, Q. J. Inebriates 19:132-141. Seltzer, C. C., 1974, Effect of smoking on blood pressure, Am. Heart J. 87(5):558-564. Shah, J. R., Warawadeker, M. D., Deshmukh, P. A., and Phutane, P. N., 1959, Institutional survey of pulmonary tuberculosis with special reference to smoking habits, Ind. J. Med. Sci. 13:381-392. Short, J. J., Johnson, H. J., and Ley, H. A., 1939, The effects of tobacco smoking on health, J. Lab. Clin. Med. 24:586-589. Silbert, S., and Friedlander, M., 1931, Studies in thrombo-angiitis obliterans (Buerger). VII. The basal metabolism, JAMA 96:1857-1858. Spring, B., 1984, Recent research on the behavioral effects of tryptophan and carbohydrate, Nutr. Health 3:55-68. Spring, B., Mailer, 0., Wurtman, J., Digman, L., and Cozolino, L., 1982/1983, Effects of protein and carbohydrate meals on mood and performance: Interactions with sex and age, J. Psychiatr. Res. 17 ( 2 ):155-167. Spring, B. J., Lieberman, H., Swope, G., and Garfield, G., 1984, Effects of carbohydrates on mood and behavior, Presented at Diet and Behavior: A Multidisciplinary Evaluation, Arlington, VA, Nov. 27-29, 1984. Stamford, B. A., Matter, S., Fell, R. D., Sady, S., Cresanta, M. K., and Papanck, P., 1984a, Cigarette smoking, physical activity, and alcohol consumption: Relationship to blood lipids and lipoproteins in premenopausal females, Metabolism: Clin. Exp. 33:585-590. Stamford, B. A., Matter, S., Fell, R. D., Sady, S., Papanek, P., and Cresanta, M., 1984b, Cigarette smoking, exercise, and high density lipoprotcin cholesterol, Atherosclerosis 52(l):73-83. Stamford, B. A., Matter, S., Fell, R. D., and Papanck, P., 1986, Effects of smoking cessation on weight gain, metabolic rate, caloric consumption, and blood lipids, Am. J. Clin. Nuir. 43(4):486-494. Stephens, T., and Pederson, L., 1983, Smoking, physical activity and health: Findings from the Canada Fitness Survey, in: Proceedings of tlre World Conference on Smoking and Health, Volume I (F. Forbes, R. C. Frecker, and D. Nostbakken, eds.), pp. 217-223, Canadian Council on Smoking and Health, Ottawa. Stitzer, M. L., and Gross, J., 1988, Nicotine replacement: Effect on post cessation withdrawal symp- toms and weight gain, in: Problems of Drug Dependence, NLDA Research Monograph No. 8 1 (L. Harris, cd.), pp. 53-58, U.S. Govt. Printing Office, Washington, DC.
Page 43: rcy63d00
Sutherland, W. H. F., Temple, W. A., Nye, E. R., and Herbison, G. P., 1980, Adiposity, lipids, alcohol consumption, smoking, and gender, Am. J. Clin. Nutr. 33(12):2581-2587. Taylor, C. K., 1910, The boy and the cigarette, Psvchol. Clin. Philadelphia 1:54-55. Theeuwes, F., and Yum, S. 1., 1977, Principles of the design and operation of generic osmotic pumps for the delivery of semisolid or liquid drug fortYl'tr7atioiis; Ann. Biomed. Eng. 4:343-353. Tuomilehto, J., Jalkanen, L., Salonen, J. T., and Nissinen, A., 1985, Factors associated with changes in body weight during a five-year follow-up of a population with high blood pressure, Scand. J. Soc. Med. 13(4):173-180. Turley, F. C., and Harrison, T. R., 1932, Respiratory measurements as affected by smoking and by athletics, Am. J. Med. Sci. 183:702-707. USDHEW, 1979, Smoking and Health: A Report of the Surgeon General, Department of Health, Education, and Welfare, Public Health Service, Office on Smoking and Health, DHEW Publ. No. (PHS) 79-50066. USDHHS, 1982, The Health Consequences of Smoking: Cancer. A Report of the Surgeon General, Department of Health and Human Services, Public Health Service, Office on Smoking and Health, DHHS Publ. No. (PHS) 82-50179. USDHHS, 1983, The Health Consequences of Smoking: Cardiovascular Disease. A Report of the Surgeon General, Department of Health and Human Services, Public Health Service, Office on Smoking and Health, DHHS Publ. No. (PHS) 84-50204. USDHHS, 1984, The Health Consequences of Smoking: Chronic Obstructive Lung 1)isease. A Report of the Surgeon General, Department of Health and Human Services, Public Health Service, Office on Smoking and Health, DHHS Publ. No. 84-50205. USDHHS, 1986, The Health Consequences of Using Srnokeless Tobacco. A Report of the Advisory Corrunittee to the Surgeon General, NIH Publ. No. 86-2874. USDHHS, 1988, The Health Consequences of Smoking: Nicotine Addiction. A Report of the Surgeon General, Department of Health and Human Services, Public Health Service, Office on Smoking and Health, DHHS Publ. No. (CDC) 88-8406. Vandenbroucke, J. P., Mauritz, B. J., DeBruin, A., Verheesen, J. H. H., Van der Heide-Wessel, C., and Vander Heide, R. M., 1984, Weight, smoking, and mortality, JAMA 252:2859-2860. Wager-Srdar, S. A., Levine, A. S., Morley, J. E., Hoidal, J. R., and Niewochner, D. E., 1984, Effects of cigarette smoke and nicotine on feeding and energy, Phpsiol. Behav. 32(3):389-395. Waller, R. E., and Brooks, A. G. F., 1972, Heights and weights of men visiting a public health exhibition, Br. J. Prev. Soc. Med. 26(3):I80-185. Wehner, A. P., Olson, R. J., and Busch, R. H., 1976, Increased life span and decreased weight in hamsters exposed to cigarette smoke, Arch. Environ. Heahh 31:146-153. Wellman, P. J., Marnion, M. M., Reich, S., and Ruddle, J., 1986, Effects of nicotine on body weight, food intake, and brown adipose tissue therntogenesis, Picarmacol. Biochem. Behar. 24(6):1605- 1609. Wennmalm, A., 1982, Effect of cigarette smoking on basal and carbon dioxide stimulated cerebral blood flow in man, Clin. 1'ln:ciol. 2(6):529-535. Whichelow, M. J., Golding, J. F., and Treasure, F. P., 1988, Comparison of some dietary habits of smokers and non-smokers, Br. J. Addiction 83:295-304. Williamson, D. F., Forman, M. R., Binkin, N. J., Gentry, E. M., Remington, P. L., and Trowbridge. F. L., 1987, Alcohol and body weight in United States adults, Am. J. Public Health 77(10):1324- 1330. Wright, S., 1846, Experiments on the physiological action of tobacco, London M. Ga:. N.S. 92:592- 593. Wurtman, J. J., and Wurtman, R. J., 1979, I)rugs that enhance central serotoninergic transmission diminish elective carbohydrate consumption by rats, Life Sci. 24:895-904. Wurlntan, R. J., 1982, Nutrients that modify brain function, Sci. Am. 246:50-59. Wurtm.m, R. J., 1987, Dietary treatments that affect brain neurotransntitters, Ann. NY Acad. Se•i. 499: 179-190. Wurtman, R. J., Hefti, F., and Mclanted, E., 1981, Precursor control of neurotransmitter synthesis. 1'harmacol. Rev. 32:315-335. x lti 1 /l~) ~ir.a:t.I'tls, 1`t 1 h I 1 S~y' `1 ~ C 2_t ~ A PclL~ 1 A .. ~ . L- S, `~ ~ 1- I 5 G N , r-i, .. : i
Page 44: rcy63d00
INVERSE RELATIONSHIP BETWEEN TOBACCO USE AND BODY WEIGHT 315 Youne, P. T., 1936, Motivation of Behavior, Wiley, New York. \'oung. P. T., 1955, The role of hedonic processes in motivation, in: Nebraska Symposiurn on Moti- ration (NI. R. Jones, ed.), pp. 193-238, University of Nebraska Press, Lincoln. Yung, 1.., Gordis, E., and Holt, J., 1983, Dietary choices and likelihood of abstinence among alcoholic patients in an out-patient clinic, Drug Alcohol Dependence 12:355. 7.einer-Henriksen, T., 1976, Rokevaner i den Norske befolking [Smoking habits of the Norwegian population], Tiddskr. Norske Laegeforening 96(11):617-620. Zuckerman, M., and Lubin, B., 1965, Manual for the Multiple Affect Adjective Checklist, Educational and Industrial Testing Service, San Diego, CA.

Text Control

Highlight Text:

OCR Text Alignment:

Image Control

Image Rotation:

Image Size: