Tobacco Documents Online

.1 I I I I I I I I I I I I I I I I r Kari-Olov Fagerstrom Ufteroker Flospitol. Uppsda, Sweden Tolerance, withdrawal and dependence on tobacco and smoking termination Introduction This paper describes the development of a questionnaire for measuring physical dependence on tobacco smoking. Data will be reported on how the instrument was validated and how it relates to phenomena such as smoking cessation, withdrawal symptoms and regulation of nicotine intake. No attempt is made to disentangle or differentiate between physical and psychological dependence. Con- sequently when dependence in the pharmacological sense is used it does not mean that psychological influences are totally ruled out. Tobacco use in the form of cigarette smoking has been widespread in industrialized countries since World War II. Despite a decline in the proportions of smokers in many Western countries during recent years, the sheer number of smokers remains high. In the USA it is estimated that in 1980 32 per cent of the adult popula- tion were smokers (US Dept HHS, 1981) and in Sweden in 1980 the figure was 29 per cent (NTS, 1981). Today there is little doubt that smoking represents a health risk. In a country the size of Sweden, with only 8 million inhabitants, the estimated annual cost to society is S250 million (US) (Jonsson, 1980). Estimate of the total annual cost for the USA is approximately S18 billion (US) (Luce & Schweitzer, 1977). Therefore it is hardly surprising that the WHO (1975) concluded that the control of cigarette smoking could do more to prolong life and improve health in industrialized countries than any other single action in the entire field of preventive medicine. Ongoing campaigns in many countries to eliminate or late,•national Review, oof Applied Psychology (SAGE. London. Beverly Hills and New Deihi). Vol. 32k1983). 29-52 I

I I I I I I I I I I I I I I I I 30 Karl-Olov FagtrstrfSm reduce tobacco smoking are not producing striking results. Among physicians, over 20 per cent continue to smoke despite their better knowledge. An exception is the USA where the figure is around 5-10 per cent. The prospects of abolishing or at least strongly diminishing tobacco use in the rest of the population through infor- mation alone would seem remote. Difficulties encountered in in- fluencing users of tobacco may be attributable to the addicting pro- perties of the substance, the social acceptability of the habit and the psychological needs fulfilled by smoking. The social acceptability guarantees that, under normal conditions, smokers are not forced into antisocial behaviors for maintaining their smoking as are for example, narcotic addicts dependent upon morphine. However, when the supply of tobacco is restricted as it was in Germany after World War II, smokers engaged in extreme beh,aviors to satisfy their need for tobacco. Despite conditions of extreme deprivation and in situations where food rations were under 1000 calories per day smokers still bartered edibles for cigarettes. The craving for tobacco, was so overwhelming that individuals were willing to engage in prostitution or larceny to procure goods to trade for cigarettes (Blair, 1979). Tobacco usage seems to flourish in societies which have been ex- posed to it unless it is specifically prohibited for religious reasons. The use of tobacco is readily adopted by cultures that are ignorant of the health risks and without social pressure for or against smok- ing (Damon, 1973). In 1974, 278 British opiate users were asked to rank a number of drugs with respect to greatest personal need. Cigarettes were rated above heroin, methadone, amphetamine, barbiturates, LSD, cannabis, alcohol and tea or coffee (Russell, 1971). The same investigator concluded some years later that 'cigarette smoking is probably the most addictive and dependence producing form of self-administered gratification known to man' (Russell, 1974). Measuring abstinence rates from several treatment programs for smokers and heroin addicts, Hunt, Barnett and Branch (1971) found that the two groups had the same absolute and relative relapse rate. Smoking tobacco seems to meet the three criteria for addic;ion. First, tolerance develops very rapidly, within hours and days, whereafter it is relatively stable (Jones, Farrell & Herning, 1979). Second, dependence on nicotine is readily observed on self- administering rats (Hanson, Ivester & Morton, 1978) and more familiarly by humans. Third, withdrawal symptoms upon disconti-

I I I I I I I I I I I I I Tobacco dependence and smoking termination 31 nuing tobacco usage are common (Shiffman, 1979). Clinical experience with individuals seeking help to stop smoking indicates that dcgree of addiction varies greatly. Those seeking help are probably more addicted, unable to quit on their own and thus amenable to help from other sources. At one extreme, clinicians find individuals who can abstain from tobacco for several days without undue physical or mental discomfort. These individuals often inhale rarely and smoke low nicotine brands. They vary the time intervals between cigarettes considerably and have no difficul- ty in refraining when smoking is not allowed. At one other extreme are individuals who light a cigarette before they get out of bed, ex- perience withdrawal symptoms when a limited time has passed bet- ween cigarettes and chain smoke the first two or three cigarettes of the day. These individuals can seldom sit through a performance or meeting without leaving for a cigarette, operating on a relatively fixed interval schedule and preferring high nicotine cigarettes. Smokers differ not only in the behavior they exhibit but also in the kinds of difficulties they encounter when they quit. The dif- ferences depicted above are most readily conceptualized as dif- ferences in tolerance and dependence on tobacco (more specifically to nicotine). Several researchers have tried to explain the differences in smok- ing behavior among individuals. The two most well-known smok- ing >'ypologies ba_sed on factor analysis of smoking related ques- tionnaires are Tomkins' 'Reasons for Smoking' Questionnaire and Russell's `Classification of Smoking by Motives' (Tomkins, 1966, Ikard, Green and Horn, 1969 and Russell, 1974). The factors found by Tomkins, Ikard, Green and Horn were called smoking for (a) pleasurable relaxation, (b) reducing negative affect, (c) stimulation, (d) sensorimotor manipulation, (e) habit, and (f) addiction. Russell, Peto and Patel's typology uses six fac- tors: (a) stimulation smoking, (b) indulgent smoking, (c) psycho- logical smoking, (d) sensorimotor smoking, (e) addictive smoking and (f) automatic smoking (Russell, 1974). Russell et al. (1975, 1976) emphasize the addiction component more than Tomkins and his associates. Despite the fact that many investigators regard tobacco smoking as addictive and dependence producing and that clinicians are daily confronted with smokers' attempts to break the habit, no systematic attempt has been made to develop a measure- ment for nicotine dependence. Such a measurement should be valuable in smoking cessation I

I I I I I I I I I I I I I I I 32 Karl-Olov Fa;erstrOm outcome research and nicotine regulation studies, since the concept of dependence seems inherently related to both cessation outcome and nicotine regulation. In both types of studies there is con- siderable variation in outcome that is partly attributable to in- dividual differences. With dependence under control, much of what earlier has been termed error variance could perhaps be ex- plained. It is also assumed that a measure of pharmacological dependence could be used as a diagnostic tool for choosing the most ap- propriate treatment strategy for smoking cessation. Me2surinQ degree of physical dependence on tobacco smokinR. A basic assumption serves as a starting point for a discussion of physical dependence: all regular smokers are addicred, whether it be psychological or physiological addiction. All are probably psychologically dependent upon cigarettes but some could also be called true addicts in the physiological/medical sense of the term. The two concepts are by no means simple to untangle since they cannot be measured directly. per se. What wou,u cnaractertzc a true cigarette/tobacco addict? It is feasible to think that physical dependence has to do with: (a) how often the drug is used (number of cigarettes/day) (b) the strength of the dose (the level of nicotine delivery of the brand) (c) the effective utilization of the drug (e.g. through regulation of inhalation) (d) smoking immediately after awakening in the morning and at a higher rate (because of nicotine's short half-life, the plasma level is very low) (e) rating the first cigarette in the morning as the best one (because of its power to alleviate withdrawal symptoms) (f) more internal stimulus control relative to external control, manifesting itself in difficulties to refrain f*'om smoking where it is prohibited e.g. in churches (as the blood nicotine level may become too low). From these six major assumptions eight questions, called the Tolerance Questionnaire (Fagerstrom, 1978), were constructed. 1. How soon after you wake up do you smoke your first cigarette? I

I I I I I I I I I I I I I I Tobacco dependence and smoking termination 33 2. Do you find it difficult to refrain from smoking in places where it is forbidden, e.g. in church, at the library etc? 3. Which of all the cigarettes you smoke in a day is the most satis- fying? 4. How many cigarettes a day do you smoke? 5. Do you smoke more during the morning than during the rest of the day? 6. Do you smoke if you are so ill that you are in bed most of the day? 7. What brand do you smoke? 8. How often do you inhale? The questions are scored so that higher scores indicate greater ad- diction. For details of the scoring consult Fagerstrom, 1978. The questionnaire has a range of 0-11 points with zero indicating minimum physical dependence and eleven points maximum physical dependence. The questions are ordered according to their loadings in a factor analysis where the dependence factor was clear- ly demonstrated. The first six questions had loadings in the factor of at least 0.60. Validation of the questionnaire Withdrawal as change in body temperature Validation of the questionnaire was first studied in relation to a withdrawal response: body temperature. It was assumed that the well-known changes in peripheral temperature as an effect of smoking (e.g. Auge, 1973; Larson. Haag & Silvette, 1961) in some way corresponds with internal body temperature. Body temperature has also been found to be reac•ive in the initial stages of abstinence to drugs such as opiates and barbiturates (Liaynert, 1968). Method. To investigate whether temperature changes could be related to degree of dependence as measured by the Tolerance Questionnaire (TQ) twenty-six subjects seeking help at a smoking withdrawal clinic were recruited. The TQ was mailed to the smokers, to be filled in before the first visit to the clinic. Body temperature was recorded within a week before and within two days after complete smoking termination. Temperature was measured orally during visits to the clinic. Variations due to time of I

I I I I I I I I I I 34 Karl-Otov FcRersrrom day and menstrual cycle were taken into consideration. Results. As an effect of smoking termination half of the twenty- six smokers showed increase and the other half showed decrease in body temperature. However, when the withdrawal response (defin- ed as body temperature before and after smoking termination) was calculated and correlated with the questionnaire, a correlation of -0.55 (p<.01) was obtained. This negative relationship means that the body temperature of the more physically dependent smokers showed an increase. Heart rate increase as acquired tolerance In the second experiment within the same study, acquired tolerance (Kalant, LeBlanc & Gibbins, 1971) among regular smokers was employed as an indicator of physical dependence. Autonomic ac- tivity operationalized as heart rate increase was used as the depen- dent variable since many researchers have documented increases in heart rate after the smoking of a cigarette (e.g. Herxheimer, Grif- fiths, Hamilton & Wakefield, 1967). Method. Nineteen regular smokers participated in this study. The. subjects began the experiment by sitting in a chair while a stable heart rate was established. For measuring heart ra,e a detec- tor was attached to the subject's chest. Once heart rate was stable, the smoker proceeded to smoke a cigarette with an interpuff inter- val of 15 seconds. Observations were made to ensure that the sub- jects inhaled and took puffs of approximately equal deFth. Heart rate increase was defined as the peak heart rate value recorded sub- tracted from the baseline heart rate. Results. The correlation coefficient between heart and the TQ score was -0.69 (p<.01), which means that the more physically dependent smokers had a smaller increase in heart rate than the less physically dependent smokers. The increase varied between 3 and 33 beats/min, the average increase being 15.2 beats/min. As in ex- periment 1, the daily cigarette consumption can be a third variable interfering with the above correlation. When the number of cigaret- tes smoked per day was partialled out, a slight attenuation oi the correlations in both experiments was found, however still within the one per cent level of significance. I

I I I I I I I I I I I Tobacco dependence cnd smokinR termination 35 Discussion The results cited above indicate that the TQ seems to measure something which appears related to a genuine addiction with its triad of tolerance, dependence and withdrawal. In the experiments, tolerance and physical withdrawal were measured and found to vary according to the answers to the eight smoking behavior ques- tions. Based on the two experiments described it seems that the TQ has a certain validity for measuring physical dependence. The utility of the questionnaire, however, remains to be demonstrated in predic- ting withdrawal symptoms, accounting for variations in smoking cessation outcome and, most essential, in treatment studies where the effects of therapy components are evaluated separately for high and low dependent subjects. Physical withdrawal responses 2nd success in smoking cessation In the next study concerning quitters' physical withdrawal responses (Fagerstrbm, 1980) the likelihood of success in quitting and reduction of the withdrawal responses with the help of a nicotine substitute were studied in relation to physical dependence. Physical wirhdrawal The majority of smokers quit with minimal physical withdrawal. symptoms. However, all experience some psychological changes, •-such as increased irritation, fluctuation in mood, aggressiveness, increased anxiety, indifference, diminished stress tolerance and craving for tobacco. In the study to be reported below, the aim was to isolate the physical symptoms of tobacco withdrawal. The distinction between physical and psychic responses is somewhat ar- tificial and by no means simple to delineate. Most of the responses recorded here are probably consequences of a drop in blood pressure that often takes place during the first two we-Cks (headache, dizziness, sweating, inability to concentrate). The other symptoms recorded were constipation (probably due to the absence of the gastrocolic reflex caused by cigarette smoking) and creeping sensations in or beneath the skin (due to spasmodic relaYation of I

I t I I I I I I I I I I I I I I I 36 Xarl-Olov Fagersrr6m the smooth musculature in the peripheral vessels). Two weeks after quitting, ex-smokers were given a questionniare in which they were asked to rate their abstinence responses. On each response they had to choose an alternative from zero (not experienced at all) to three (experienced very severly). When the total sum of these ratings was calculated and correlated with their TQ score a positive correlation of 0.40 (p<.01) was found, i.e. the more dependent the subject, the more withdrawal responses s/he is likely to experience. To ascertain whether these withdrawal symptoms were related to absence of nicotine and consequently could be reduced by ad- ministering nicotine, thirty smokers who wanted to quit were given a supply of nicotine chewing gum (Nicorette,) and com- pared with thirty smokers who received no nicotine substitute. These two groups differed significantly (p<.05) in terms of physical withdrawal symptoms. Those having access to the nicotine substitute experienced fewer symptoms. There also seemed to be a relationship between the quantity of nicotine gum chewed and degree of physical withdrawal. The more gum consumed, the less withdrawal symptoms recorded (p<.05). No placebo group was compared with these smokers, so these findings can only be sug- gestive. Likelihood of success in quitting The outcome of a smoking cessation attempt is influenced by a vast number of factors, for example how does physical dependence on tobacco smoking influence cessation? In order to investigate this question 130 patients at the smoking cessation clinic were classified as smokers or ex-smokers three months after quitting. The average scores on the TQ for the successful and unsuccessful quitters were 6.2 and 7.0 (p<.05) respectively, which implies that the unsuc- cessful subjects were more physically dependent than their suc- cessful counterparts. This difference in outcome for smokers with different degrees of dependence has been replicated many times at the clinic, and usually with only 50-75 subjects needed to establish a significant difference. The TQ thus sezm to be a good predictor of outcome. Since the highly dependent smokers have a substantially reduced probability of succeeding and since the common abstinence rates at twelve months for most smoking cessation pro- grams are as low as 20 per cent (Leventhal & Clearly, 1980) one I

I I I I I I I I I I I I I I Tobacco dependence and smokinR terminarion 37 wonders how low the percentage abstinent is for the highly depen- dent subjects. Conclusions and implications These results suggest that: (a) the degree of dependence determines or at least covaries with the more physical withdrawal responses that smokers may encounter, (b) nicotine is likely to be a crucial factor in the etiology of the withdrawal responses, (c) nicotine in the form of a chewing gum may be helpful in reducing the abstinence responses, (d) unsuccessful quitters are more physically dependent on cigarettes and probably require other intervention strategies than less dependent smokers. Nicotine chewing gum could well be such an intervention for smokers who then could quit with a two- step program. The first steps aims at stopping the smoking and the second step at gradually reducing the amount of nicotine used as the strength of the smoking habit diminishes. A placebo controlled evaluation of the effectiveness of a nicotine gum in smoking cessation In order to test the assumption above, i.e. that a nicotine substitute in the form of a chewing gum could be a valuable aid in smoking cessation, especially for highly dependent smokers, a double-blind controlled study was carried out at the smoking cessation clinic (Fagerstrdm, in press a). Method • Subjects and design. One hundred consecutive patients from the clinic's waiting list were randomly assigned to the experimental (nicotine chewing gum and psychological treatment) or control (placebo gum and psychological treatment) groups. Fifty-nine of the one hundred smokers were women. The patients were seen individually by the author and given in- dividualized counseling, averaging 7.7 sessions per patient. The sub- jects in both the nicotine and placebo gum groups were told that they would have access to chewing gum containing nicotine, which I

I I I I I I I I I I I I I I I 38 Kar!•Olov Fogerstrom TABLE 1 A comprehensive smoking cessation program .4ssessment Ten days recording of tar and nicotine intake Medical check-up Carbon monoxide concentrations pre- and post-cessation Review of motives Feedback to the medical motives, e.g. demonstration of decreased circulation contingent on smoking Recording of weight Treatment components General medical and health informauon Nonspecific support, encouragement and reinforcement ~ Warm accepting atmosphere Patient decides quitting day Education in self-control techniques Administration of chewing gum (nicotine and placebo) Sensitization of how smokers relapse Personal contact 4-6 months with 6-l5 sessions After 6 months post cards should be mailed to the clinic Opportunity to smoke at the clinic whenever desired Ocrarionally used components Rapid and aversive smoking Stimulus control Rardy used components Hypnosis Relaxation Covert sensitization Contracting Furtdamental premrses Complete absttnence should be reached at least withtn 20 days Highly nicotine dependent subjects cannot take occasional cigarette Gradual reduction is not a serious alternative to 'coid turkey' High deyree of individualizatton would be administered from the first session. The double-blind code was broken by the author when the subject had ceased to use the chewing gum, never earlier than three months after the in- dividual stopped smoking. Treatment. In addition to the placebo or nicotine gum, patients also took part in a comprehensive psychological smoking cessation