Tobacco Documents Online

I. ]:.1 1.2 Z.3 CONTENTS Introduction Statistics Psychosocial Factors Individual Differences °.. o . ..~ • ° . ." - °" °. II. III. III.l. III.2. III.3. III.4. III.5. State of the Organism Etiology of Tobacco Smoking Genetics Morphology Personality Neurophysiology Prenatal Mechanisms IV. Empirical Integration IV.I. Need for a Longitudinal Study Ve V.1. V.2. V.3. V.4. Tobacco and Stress Reduction: Affect Performance Psychophysiology Neurophsyiology An Operational Framework VI. Summary, Rationale and Objective of Study VII. Significance of the Study VIII Methodology VIII.l.Subjects VIII.2.General Design and Procedures VIII.3.Psychophysiological Test Battery VIII.4.Statistical Analyses Bibliography i J ~o C~ CD ~0 • • . . ° BatCo document for PFSFC 1 March 1999

IDENTIFYING PSYCHOPHYSIOLOGICAL PREDICTORS OF TOBACCO USE IN CHILDREN: A FI~-YEAR PROSPECTIVE" LONGITUDINAL STUDY Verner J. Knott, D. Phil. I- INTRODUCTION i. Statistics The recent report of smoking habits of Canadian school children by the Department of National Health and Welfare* (1980) has shown that students start experimenUing with cigarettes at an early age. The major conclusions regard/ng prevalence of the habit were as follows: - By age 12, one half of Canadian school children have at least tried smoking; - Regular smoking is established in the early teens by many Canadian students; - School aged boys experin~nt with cigarettes earlier than girls, but girls begin r?gular smoking earlier than boys. From the early teens, a hlgher proportion of girls than boys smoke daily; - By age 14, 15% of boys and 20% of girls are daily smokers and by age 17 these figures have increased to 27% of boys and 30% of girls. No significant change occurs in the proportion of students who report daily smoking beyond this age. These figures are remarkably similar to the studies reviewed by Russell (1971) in which it was found that of those children who smoked more than one or two casual cigarettes before the age of 19, 80% went on to become regular smokers as adults. It is only the teenager who never attempts, or who has attempted no more than once and decided that he dislikes it and will not take it up, who has much chance of being a non-smoking adult. The matter is largely settled by the age of 19; if a person still is a non-smoker at this age he is unlikely to take it up. ~a~E: see Sra~n, ~erry and Forbes [1980) for reference t~ ~ data -° 0 r,O 0", 0 ,..0 BatCo document for PFSFC 1 March 1999

-2- Q 2. Psychosocial Factors But to establish the prevalence of cigarette smoking in children and adolescents does not explain how it is contracted. There is a general consensus that the motivating factors mediating the onset of the smoking habit are distinct from those factors maintaining the habit. This has been discussed by Russell (1971) and the argument will not be repeated here other than to say that after initiation to smoking has been me.diated by various psychosocial motives, the pharmacological effects of nicotine take over as prime reinforcers. There is strong evidence that starting to smoke is related to social factors, particularly the influence of parents, siblings and peers (Mausner and Platt, 1971). The effect of the behavior of parents, peers and siblings on childhood smoking is presumably mediated partly through increased exposure to cigarettes, partly through imitation and example and partly through psychological and social pressures to smoke. Evidence for social factors such as parent's attitude and health education is to a great extent derived from cross-sectional correlational studies which give little insight into how these influences act on the school-going child. Evans, Henderson, Hill and Raines' (1979) review of the psychological factors reveals that smoking An children is related to lower levels of perceived (and real) academic achievement, to rebelliousness, anti-authority behavior, low self-esteem and that children who begin to smoke at an earlier age show a higher rate of sensation- seeking behaviors than non-smoking peers. 3. Individual Differences Although it is generally stated that social and psychological factors are responsible for the initiation of smoking, it is clear that no sharp line divides the biological from the psychological and sociological. The kinds of inner emotions that people experience in relation to environmental conditions and events is at the same time an expression of both psychological and biological factors. These • emotions and the environmental situations that are perceived as rewarding or aversive will determine, at least in part, a wide variety of behaviors including one's response to cigarette smoke (Jaffe and Jarvik, 1978). Individual differences in personality- and particularly in orientation towards the kind of psychological rewards offered by smoking are considered to be of paramount importance O o P~D C7~ Co --.j O G~ O BatCo document for PFSFC 1 March 1999

-3- in determining both smoker/non-smoker status and type of smoking behavior. This is illustrated with respect to early experimentation with cigarettes. The reasons most people give for their first experiments with smoking are ~riosity, conformity, bravado or to appear grown-up (Horn, Courts, Taylor and Solomon, 1959). The first cigarette is almost invariably unpleasant. Nearly everyone has experienced the shock of the first inhalation, often accompanied by gagging and nauseousness - a physiological response which presumably has a pharmacological basis. On the other hand, nearly everyone who has continued smoking more cigarettes can recall the pleasantness of the light-headedness which followed inhalation, a sensory experience that could be made to recur following a brief time lapse between cigarettes. Tolerance is said to develop to the unpleasant side-effects and skill is quickly acquired to limit the intake of smoke to a comfortable level, thus lowering the threshold for further attempts. Herein lies a possiblQ cause of the virtual inevitability of escalation after only a few cigarettes (Russell, 1971). With curiosity satisfied by the first cigarette, the act is likely to be repeated only if the physical discomfort is outweighed by the rewards of smoking. If these motives are sufficient to cause smoking to be repeated in the face of unpleasant side-effects, there is little chance that smoking will not continue as these side effects rapidly disappear. What then is the nature of these rewards and what are their basis? There is a growing consensus that organismic-dispositional factors contribute to the development of the smoking habit but as to what extent such organismic factors can be attributed to genetic biological influences as opposed to acquired influences is as yet unclear {Battig, 1980). The emphasis on organismic trait and state factors in influencing smoking behavior has been expressed in varying degrees by a number of reseachers. In attempting to extract meaning from their highly variable data on .students, McArthur, Waldron and Dickinson (1958) hypothesized "...that starting to smoke J~; largely brought about by one's social environment but that reactions to smoking ... seem to depend in good part on the personal needs that the newly established habit is able to gratify (p. 272)." Seltzer {1962) paraphrased those observations in biogenetic terms: "Rather than a superficial habit overlaid indiscriminate- ly upon various persons, smoking appears to be a response tO a wide variety of personality and behavioral characteristics whlc~ have their origin, in part, in the biologic and genetic . . O O Cr~ .....a, BatCo document for PFSFC 1 March 1999

1 -4- make-up of the individual (p. 43)." Dunn (1973) made similar references to organismic variables with specific references to early experimentation with cigarettes: "...it is likely that we shall ultimately find that the critical mechanisms involved, in smoking require the synergistic presence of some other factor, such as anxiety or possibly some constitutionally determined reactivity. The observation that only about half of those w~o try smoking cigarettes go on to take up the habit certainly suggests some kind of interaction process. Smoking apparently is not sufficiently pleasurable among the disinterested half of the tryers for the aversiveness of smoking to be overridden. Smoke is smoke - whether its inhalation is pleasurable is dependent upon what the individual brings to the situation ... Whether or not the total experience is judged by the smoker as positive or negative might in time prove to be associated with biochemical, endocrinological, or neurophysiological variables which are either constitutionally fixed or shaped by experience prior to initial experimentation with tobacco smoke (p. 101-102)." If- STATE OF THE ORGANISM The average pack-a-day smoker takes 8-10 puffs per cigarette and absorbs approximately 50-150 ug of nicotine per puff. Each dose of nicotine reaches the brain within 7 seconds and exerts widespread and varied central and peripheral nervous system effects (both stimulation and depression, all of which are potential reinforcers), via its capacity to affect the actions of and release of important neurochemical transmitters (Russell, 1976). For the average smoker the behavior is reinforced approximately 50-70 thousand times a year and this consumption level would tend to suggest that organismic factors of a markedly compelling nature are operative in the inception and persistence of smoking behavior. While environmental cues and contingency factors play a crucial role in tobacco usage, a comprehensive analysis of smoking also must take into consideration the effects of tobacco in relation to organismic variables. The mode of interaction between a pharmacological agent and ongoing organismic state is of critical importance therapeutic- ally but its study has also proved valuable in furthering an understanding of target physiological systems by the known actions of such psychoactive drugs (Irwin, 1968). A similar approach can be taken in tobacco research, namely to determine tobacco's effects on the organism, taking into consideration ongoing states, so as to elucidate possible mechanisms and sought for effects by the individual. Nicotine, as a chemical, does not affect behavior directly but rather interacts with other chemicals at a cellular level to produce changes in O O'- CO CD F~D ! o. BatCo document for PFSFC 1 March 1999

-5- tissue, organ and systemic functioning. The resulting alteration An physiological state then determines the limits and manner by which the individual copes and interacts with his environment. Thus the mode of action of nicotine at the biochemical and physiological levels may provide useful information of its possible dissimilar effects in smokers and non-smokers that may result in differential behavioral patterns of consumption. The issue is raised as to whether smokers are unique in their reasons for selection of and response to tobacco and hence consume this substance in an effort at modifying or manipulating a target system (or systems). Their consumption therefore would have ultimately behavioral consequences that alter and possibly increase coping capabilities (MAlls, 1978). Such an interpretation of tobacco usage has an obvious teleological fla~or insofar as it presupposes that smoking serves at least a potentially adaptive purpose. As will be elaborated upon more fully later, there is some indication that this may indeed be the case for some smokers. However, before proceeding further, a conceptual framework is proposed so as to allow the reader to establish a context or perspective for the position to be outlined. If, as pointed out earlier, the organismic state is an important determinan~ of tobacco consumption, then it remains to be elucidated as to what target physiological system or systems the person seeks to modify and toward what altered state the individual is striving. Irwin (1968) summarised some of the more salient organismic conditions from psychopharmacological studies. All of the factors - including wakefulness, arousal, activity, endurance, biosocial drives, set, responsiveness to stimuli, information processing and autonomic, neurophsyiological and endocrine functioning have been investigated in tobacco research and have been suggested as reasons for consumption at one time or another (Dunn, 1973). A person who voluntarily takes a drug often does so with the intention of altering one or more of the organismic .variables mentioned above.The point to be made here is that the motivation for and effect sought can vary between individuals and serve diverse functions in a given person in different situations. This fact is of paramount importance in the study of smoking motivation. First, it emphasizes multiple causality An the onset of the habit and illustrates how consideration of organismi= states precludes unidimensional etiological CD ~O <Do O Cr~ "° . BatCo document for PFSFC 1 March 1999

-6- mechanimsms as explanations for all smokers. A second point is that in addition to motivational needs and sought for~ altered organismic states that tobacco can induce, another factor for tobacco use may be intrinsic physiological disturbance that is rectified by tobacco. Thus a pre-smoker person who is defective in one aspect of physiological functioning may seek tobacco with essentially medical intent, much as a person with adrenocortical insufficiency craves salt (Wilkens and Richter, 1940). Thus, the study of tobacco consumption must incorporate into a comprehensive explanatory system in understanding of the state of the organism which under appropriate environmental circusmtances (cues and contingencies) leads to experimentation and repeated use of tobacco by an individual for its perceived and/or real effects in modifying coping capacity through its mediating action on target physiological systems. III- ETIOLOGY OF TOBACCO SMOKING From the preceeding discussion it is apparent that the motivation for using pharmacological substance can vary between individuals, and that a given chemical may also possess unique effects for certain individuals, whose organismic state dispositionally distinguishes them from others. Among the results obtained in animal research in the last few years there has been numerous indications that psychogenetic inheritance can influence the mode of action of pharmacological substances affecting the central neuropsychological systems and of substances having other kinds of effects as well (Broadhurst, 1977). Although Battig (1980) has cited animal evidence indicating response variability to nicotine as a function of psychogenetic strains, the extent such a disposition in man can be observed and attributed to genetic biological factors is open to question. Of the limit-less number of organismic variables that can lead to physiological and behavioral disruption and thus possible use of tobacco, the concept of disturbed arousal has probably received the most attention (Thornton, 1978; Remond and Izard, 1979). Disturbances in arousal mechanisms have been implicated in a variety of ~sychopathological conditions such as hyperactivity (Hastings and Barkley, 1978), anxiety and depression (Lader, 1975), sociopathy (Quay, 1965) and schizophrenia (Venables, 1977). Arousal disturbance has been theorized to be an etiological factor in children who as adults are at risk for these disorders. The question is raised as to whether tobacco may also have an arousal syndrome as an etiological substrate. The tentative hypothesis advanced is that within a population of children the=e exists a subgroup whose symptoms of arousal disturbances is premorbidly extant. Such persons will exhibit higher probabi- lity of acquiring the tobacco habit on the basis that initial experimentation with oigar@ttes will result in cD "'.. CXD O BatCo document for PFSFC 1 March 1999

° -7- alterations of organismic states which will be perceived subjectively and behaviorally as reinforcing. The evidence Implicating a premorbid difference in smokers and non-smokers Is briefly reviewed below. I. Genetics Many studies that have implicated biological factors in the initiation of smoking behavior attribute the behavior to a genetic predisposition. Studies of tl~ins are among the most popular means of assessing genetic factors. Initial twin studies by Fisher (1958) showed thot monozygotic twins were more concordant in their smoking behavior than dizygotic twins, ctutn by Shields (1962) on monozygotic twins reared apart indicated frequency of concordance in smoking status which was significantly different from chance expectancy. Subsequent studies however have both supported and denied a significant genetic influence on smoking behavior (Jarvik, 1979} and an example of negative results is shown in a study by Cederlof, Friberg and Lundman (1977) in which the results of their monozygotic twin series speak strongly against the constitutional hypothesis. The presence of a substantial number of discordant twins in these studies indicates that genetic factors do not operate exclusively to determine smoking tendencies. As an example, an estimate of the amount of variance with regard to smoking in Shields' (1962) sample which can be accounted for by genetic f~ctors is somewhat less than 33%. It seems reasonable to conclude that genetic factors operate significantly, but by no means exclusively, in the tendency to smoke; and they may do so in a wide variety of modes, including personality characteristics, and social and psychological needs and values, rather than simply by producing an inborn craving or need which nicotine satisfies (Kety, 1973). This line of thought parallels Eysen~'s(1980) conclusions on his most recent twin studies: "Our data do make us question the simple-minded exploitation of the twin design and its conventional genetic interpretation in relation to smoking. In particular, although we agree that dizygotic twins are less alike than monozygotics in aspects of the smoking habit, we are not so sure that the similarity of twins for onset and consumption of tobacco ks purely genetic... All in all, therefore, the picture emerges from the whole body of data that the onset of smoking and the consumption of cigarettes are governed by both genetical and environmental factors (p. 281, 313).". CD C~D CD Cr~ BatCo document for PFSFC 1 March 1999

-8- 2. Morphology Although the above studies suggest the existence of genetic mechanisms in the development of tobacco smoking, they give no clue as to what they might be. They only indicate that the smoker differs in some significant respect from his non- smoking peers. Early morphological work by Seltzer (1963) suggests that smokers differ from non-smokers on a variety of anthropometric measurements (e.g. height, weight, head oircumference, etc.). On every single measure e~amined, smokers were found to exhibit larger mean, dimensions than the non-smokers. As the measures involved a strong genetic component, Seltzer concluded that smoking behavior may have a constitutional basis. Although this data suffer from the fact that they may have been affected in turn by the smoking habits of the subjects investigated, this is not true of such studies as Thomas and Cohen (1960) on ability to taste phenyl%hiourea (PTC), a trait which has been demonstrated to have a genetic basks; they found that heavy cigarette smokers showed a significantly higher porportion of tasters than did non- smokers. Although the functional significance of these findings ks questionable, this is not the case with Seltzer's (1967) most recent longitudinal work. Here, future smokers, compared to future non-smokers, were found to have smaller tidal air values, an increased frequency of sighs and swollows, greater respiratory rate, a somewhat higher recumbent pulse rate, more palpitations, more sinus arrhythmia, more constipation, more loss of appetite, and a greater frequency of urination. The future non-smokers, on the other hand, exhibited a consistent lack of physiological reaction to stress, suggesting that smokers are more prone to patterns of reducing anxiety which involve physiological change. This is in good agreement with the alleged tension-reducing properties of smoking (Gilbert, 1979}. 3. Personality The possible existence of a constitutional difference between smokers and non-smokers suggest a relationship between personality and smoking. Of the numerous measures available, the majority of research in the smoking area (Eysenck, 1980} has focused on three major dimensions: extraversion-introversion ~E); neuroticism-stabillty (N); and psychoticism {P). Each major dimension is a combination of intercorrelations between traits. Thus individuals with high scores E show traits such as sociability, impulsiveness, carefreeness, activity, etc. High N individuals exhibit such traits as worry, tenseness, anxiety, emotionality, etc., and high P individuals exhibit <:D r~O CO CD O~ C7~ BatCo document for PFSFC 1 March 1999

° ° ..... _2" ..... -9- such traits as emotional coldness, hostil~ty, egocentricity, la¢k of superego control, etc. There is good evidence for the geneti¢ determination of these major personality variables and the behavioral manifestations of these personality traits would seem to be mediated through various anatomical and physiological features of the organism (Eyseck, 1967). In arousal terminology, extraverts seem to be characterised by low resting levels of cortical arousal, whereas introverts have relatively high resting levels of cortical arousal; these levels are presumably mediated by the ascending reticular formation. Relative to extraverts who are hypothesized as exhibiting strong inhibitory tendencies, low cortical arousal and hypo-sensitive responsivity to stimuli, introverts are hypothesized to exhibit weak inhibitory and strong excitatory tendencies resulting in relatively high cortical arousal and hyper-sensitivity to stimuli. Differences in emotio~ality, characteristic of N, are governed by the visceral brain, a limbic system, coordinating the activity of the sympathetic and parasympathetic autonomic systems. Physiologically, high neurotic/anxiety subjects tend to respond more strongly to stimuli, show greater variability ~n response, and take a longer time to recover after the response. As far as psychoticism is concerned, the evidence suggests some degree of hormonal control, related to sex hormones An general. Physiologically, individuals with high P scores and individuals exhibiting psychopathic behavior tend to exhibit low tonic resting levels of arousal and concomitant hyper-reactivity to stimuli and stress situations. Although inconsistencies are apparent, reviews of the smoking literature are in general in agreement that smokers exhibit higher scores on E, N and P dimensions (Matarazzo and Saslow, 1960; Eysenck, 1973; 1980; Kozlowski, 1979), and on average there seems to be a positive correlation between these dimensions and the degree of smoking within the smoking population. As with other studies and other variables however, whenever differences are reported between smokers and non-smokers, it is usually impossible to determine whether the differences 'reflect consequences of smoking or possibly have a causal relationship to smoking behavior. This objection applies to a lesser extent to studies on child smokers and to studies where personality assessments were obtained prior to smoking onset. Powe~l, Stewart and Grylls (1979) examined E, P, and N dimensions on a sample of 808 middle-class children between 7 and 16 years of age. More than half of the boys and girls O ~O C~ CO O Cr~ BatCo document for PFSFC 1 March 1999