to be the relatively most effective one. There was evidence, according to Horn, that "this approach was most effective among those who smoked in emulation of their parents, and less SO among those who smoked for the more emotionally tinged reasons of compensation or rebellion." Unfortu. nately, it is not entirely clear from the description of the study how trust. worthy was the identification of the motives underlying these children's smoking. Yet, these results agree logically with the position that there is no single cause or explanation of smoking, but that smokers may start. continue, and discontinue smokin g in response to different inner needs and external influences, social and other. SUMMARY Scientific investigations into the psycho-social aspects of smoking are relatively recent and, except for a few large-scale and systematic studies, leave much to be desired from the standpoint of methods and conceptions. However, evidence from a few sound studies, and converging evidence from many studies, none of which could stand up by itself under exacting scrutiny, permit the following statements concerning the relationship between psycho- social characteristics and smoking behavior: 1. As far as is known from actual data, few children smoke before the age of 12, probably less than five percent of the boys and less than one percent of the girls. From age 12 on, however, there is a fairly regular increase in the prevalence of smoking. At the 12th grade level between 40 to 55 percent of children have been found to be smokers. By age 25, estimates of smoking prevalence run as high as 60 percent of men and 36 percent of women. There is a further increase up to 35 and 40 years after which a drop is observed. In the 65 and over age group. prevalence of smoking is only approximately 20 percent among men and 4 percent among women. 2. Smokers and non-smokers differ in a number of demographic character- istics but no single comprehensive theory to explain smoking is suggested by the demographic data taken by themselves. 3. Although smokers are different from non-smokers psychologically and socially, there are many differences among smokers and among non-smokers, so that some smokers may be like some non-smokers. 4. Smoking appears to be not one behavior but a range of psychologically diverse behaviors each of which may be induced by a different combination of factors and may serve different needs. Therefore no single explanation can suffice. 5. Social stimulation appears to play a major role in a young person's early and first experiments with smoking. 6. There is suggestive evidence that early smoking may be linked with self-esteem and status needs although the nature of this linkage is open to different interpretations. 7. No scientific evidence supports the popular hvpothesis that smoking among adolescents is an expression of rebellion against authority. 376 8. No differences in intelligence between smoking and non-smoking chil- dren have been found, but smokers are more frequent among those who fall behind in scholastic achievements. 9. No smoker personality has been established but certain personality fac- tors have been reported to be associated with smoking, among them extro- version, neuroticism, and a disproportionate prevalence of psychosomatic manifestations. 10. Stress appears to be less associated with prevalence of smoking than with fluctuations in amount of smoking. 11. The cultural milieu seems to have a strong influence, a permissive cul- tural climate tending to promote and a rejecting or outright prohibitive one to inhibit smoking. 12. Less is known about discontinuation than about beginning of smoking. although there is good evidence that it is related to the beginning of the habit, its nature, and duration. CONCLUSION The overwhelming evidence points to the conclusion that smoking-its beginning, habituation, and occasional discontinuation-is to a large extent psychologically and socially determined. This does not rule out physiological factors, especially in respect to habituation, nor the existen,ce of predisposing constitutional or hereditary factors. REFERENCES 1. A Report on Personality Factors and Smoking. Part 2. [Eysenck supv] Prepared by Mass-Observation Ltd. August 1962. 27, 24 p. 2. Bothwell, P. W. The epidemiology of cigarette smoking in rural school children. Med Offr 102: 125-32,1959. 3. Cartwright, A., Martin, F. M., Thompson, J. G. Distribution and devel- opment of smoking habits. Lancet 2: 725-7; 1959. 4. Davis, R. Cigarette smoking motivation study. Research Services, London, 1956. 5. Earp, J. R. The student who smokes. Yellow Springs, Antioch 1931. 64 P. 6. Eysenck, H. J., Tarrant, M., Woolf, M., England, L. Smoking and personality. Brit Med J 1: 1456-60, 1960. 7. Haenszel, W., Shimkin, M. B., Miller, H. P. Tobacco smoking patterns in the United States. Public Health Monog No. 45: 1-111, 1956. 8. Hammond, E. C. Report to the Surgeon General's Advisory Committee on Smoking and Health. 9. Hammond, E. C., Percy, C. Ex-smokers. N Y J Med 58: 29569,1958. 10. Heath, C. Differences between smokers and nonsmokers. AMA Arch Intern Med 101: 377-88,1958. 377 11. Horn, D. Behavioral aspects of cigarette smoking. J Chronic Dis 16. 383-95. 1963. 12. Horn. D. Modifying smoking habits in high school students. Children 7: 63-5. April l%O. 13. Horn. D.? Courts, F. A., Taylor. R. M., Solomon: E. S. Cigarette smokina among high school students. Amer Public Health 49: 1497, 1959, 14. Kissen, D. M. Emotional factors cigarette smoking and relapse in pul. monarp tuberculosis. Health Bull 18: 38-44, 1960. 15. Kissen. D. M. Psycho-social factors in cigarette smoking motivation. Med Offr 104: 365-72. 1960. 16. Lawton. M. P. Psycho-social aspects of cigarette smoking. J Health Hum Behav 3: 163-70,1962. 17. Lawton. M.. Goldman. A. Cigarette smoking and attitude toward the etiology of lung cancer. J Sot Psycho1 54: 235-48, 1961. 18. Lawton, M., Phillips, R. The relationship between excessive cigarette smoking and psychological tension. Amer J Med Sci 232: 39742. 1956. 19. Lilienfeld, A. Emotional and other selected characteristics of cigarette smokers and nonsmokers as related to epidemiological studies of lung cancer and other diseases. J Nat Cancer Inst 22: 259-82, 1959. 20. Lynn. R. M. A study of smokers and non-smokers as related to achieve- ment and various personal characteristics. [Abstract J In: Res Prog No. 464, p. 164, 1948. 21. McArthur, C. C. The personal and social psychology of smoking. In: James, G.. Rosenthal, T. eds. Tobacco and Health. Springfield, Ill., Thomas, 1961. p. 201-9. 22. McArthur. C.. Waldron, E., Dickinson, J. The psychology of smoking. J Abnorm Sot Psycho1 56: 267-75. 1958. 23. hlatarazzo, J. D.. Saslow, G. Psychological and related characteristics of smokers and non-smokers. Psycho1 Bull 57: 493-513,196O. 24. Matarazzo. R. M.. Matarazzo, J. D.. Saslow, G., Phillips, J. S. Psycho- logical test and organismic correlates of interview interaction pat- terns. J Abnorm Sot Psycho1 56: 329-38, 1958. 25. Moodie. W. Smoking. drinking, and nervousness. Lancet 2: 188-9, 1957. 26. Sackrin, S.. Conover. A. Tobacco smoking in the U.S. in relation to income. L-SDA Market Res Rep Xo. 189, 1957. 27. Salber, D.. MacMahon. 8. Cigarette smoking among high school stu- dents related to social class and parental smoking habits. Amer J Public Health 51: 1780-9, 1961. 28. Salber, E., Worcester, J. Ch ange in women's smoking patterns. Cancer. In press as of September 1963. 29. Salber. E. J., Welsh. B.. Taylor, S. V. Reasons for smoking given by secondary school children. J Health Hum Behav 4: 118-29, NO. 2, Summer 1963. 30. Salber, E. J., and others. Recreational activities and attitudes toward smoking: A study of school children. Pediatrics. In press as of Sep- tember 1963. 378 31. Salber, E. J., Goldman, E., Buka, M., Welsh, B. Smoking habits of high school students in Newton, Mass. New Eng J Med 265: 969-74,1961. 32. Salber, E. J., MacMahon, B. S mo k ing habits of high school students related to intelligence and achievement. Pediatrics 29: 780-7, 1962. 33. Schonfeld, J. Special report to the Surgeon General's Advisory Com- mittee on Smoking and Health. 34. Schubert, D. Volunteering as arousal-seeking. [Abstract] Amer Psycho1 15: 413, 1960. 35. Seltzer, C. Masculinity and smoking. Science 130: 1706-7, 1959. 36. Strachey, J., ed. Sigmund Freud three essays on the theory of sexuality. In: Inst Psycho-Analyt Lib. London, Hogarth, 1962. No. 57, p. l-130. 37. Straits, B., Sechrest, L. Further support of some findings about the characteristics of smokers and non-smokers. J Consult Psycho1 In press, 1963. 38. Thomas, C. B. Characteristics of smokers compared with non-smokers in a population of healthy young adults including observations on family history, blood pressure, heart rate, body weight, cholesterol, and certain psychologic traits. Ann Intern Med 53: 697-718, 1960. 39. Todd, G. Statistics of smoking. 2d edition. Research Paper No. 1. The Tobacco Manufacturers' Standing Committee. London, 1959. 379 Chapter 15 Morphological Constitution of Smokers Contents PHYSIQUE OF SMOKERS Page ................ 384 SOMATOTYPE CLASSIFICATION ............ 38:( MASCULINITY ..................... 383 BODY WEIGHT. .................... 384 PROSPECTZVE STUDY ................. :I85 CONCLUSION ...................... :<87 REFERENCES. ..................... X8; 382 Chapter 15 MORPHOLOGICAL CONSTITUTION OF SMOKERS PHYSIQUE OF SMOKERS Several studies deal with the relation of morphological constitution and smoking. In 1929 Diehl (2) reported a study of the physique of smokers as compared to non-smokers in a group of freshmen at the University of Minnesota. Measurements of height and weight were obtained at the time of the freshman entrance examination, and smoking habit was determined from a questionnaire item based simply on whether the student did or did not smoke. No significant differences were found in height, weight, and height/weight ratio between the 445 smokers and 441 non-smokers. How- ever, the design of the study limits the reliability of the information. SOMATOTYPE CLASSIFICATION A more satisfactory but still limited study was reported by Parnell (4) in 1951. Using Sheldon's somatotyping technique, Parnell contrasted the classifications of smokers and non-smokers of 308 Oxford undergraduates. In smokers the most frequent somatotypes were the dominant endomorphs and endomorphic mesomorphs; the least frequent was the dominant ecto- morph, with the dominant mesomorph in the middle. For the non-smokers the most frequent somatotype was the dominant ectomorph, and the meso- morphic ectomorph; the least frequent were the endomorphs and the endomorphic mesomorphs, and again the dominant mesomorphs were in the middle. MASCULINITY In 1959 Seltzer (5 j presented information on the relationship between physical masculinity and smoking in a group of 247 Harvard College students who had been followed for more than 15 years for smoking habits, as well as other information. From the smoking data, the subjects were classified into three groups, non-smokers, moderate smokers and heavier smokers. When the subjects were sophomores, they were rated with respect to a body- build complex known as the masculine component, which referred to the element of masculinity as indicated by external morphological features. In measuring this element, the more the pattern of anatomical traits tends 7 14-422 O-64-26 383 toward the extreme masculine form, the stronger is the masculine component. the greater the departure from the extreme masculine type towards th; feminine build, the weaker is the masculine component. The results of this study showed a statistically significant association between the strength of the masculine component and smoking habits. More specifically, it was found that weakness of the masculine component is significantly more frequent in smokers than in non-smokers, and most frequent in heavier smokers. Furthermore, it was indicated that the subjects with weakness of the masculine component showed a constellation of personality and behavioral traits that were, for the most part, not inconsistent with the findings of Heath (3) in his study of the differences between smokers and non-smokers, Although these findings were suggestive, they were recognized by the author as being preliminary and tentative in nature and requiring further confirma. tion. Furthermore, the series on which these results were obtained was relatively small and represented a highly selected population. BODY WEIGHT Thomas (7)) in her study of precursors of hypertension and coronary artery disease in more than 1,000 students at The Johns Hopkins University School of Medicine compared the group of non-smokers with the group of smokers for body weight among other characteristics. The group of 297 non-smokers included occasional smokers as well, and the 321 smokers in- cluded all smokers except non-smokers, occasional, ex-smokers, and unknown. Pipe, cigar, and mixed smokers were included in the smoker category. The relationship of body weight to smoking habits was analyzed on the basis of percentage of overweight and underweight calculated from standard tables. Thomas found the percentage distribution of overweight and underweight was similar for smokers and non-smokers except at the upper end of the distribution curve. There was an excess of smokers who were 30 percent or more overweight, and the subjects who were 4Q percent or more overweight were all regular smokers, The non-smokers had also a greater frequency of individuals with 10 percent or more underweight than the smokers. The difference between smokers and non-smokers with regard to this body weight classification was found to be statistically significant. The subjects were also compared for the ponderal index (height over the cube root of weight), with the smokers showing an excess of the unusually heavy body builds. In the introduction to her paper on the characteristics of smokers com- pared with non-smokers (of which the weight analysis was a part), Thomas wrote: "The finding that smokers, especially heavy smokers, have a higher mortality rate from coronary heart disease than do non-smokers makes it important to determine whether those who smoke are fundamentally different from those who do not smoke, or whether smokers and non-smokers are essentially alike. If alike, th en smokers and non-smokers may be considered as a single population with a uniform life expectancy. If, however, smokers have constitutional differences from non-smokers, the two groups might have 384 inherently different mortality rates, and one group could not serve as a control for the other in statistical studies." After detailing the significant differences noted in her data between smokers and non-smokers, with regard to history of parental hypertension, heart rate, pulse pressure, body weight, and other variables, Thomas concluded that "It cannot be determined from the present data whether those individual characteristics which are more often found among smokers than non-smokers represent true constitutional differences or are due to the effects of smoking. The differences observed in the parental histories indicate that smokers and non-smokers have a somewhat different heritage, and suggest that at least some of the variations found in individual traits may be genetic in origin." In a study of 167 adult male factory workers of Neapolitan parentage but of American birth and upbringingt Damon (11 reported on morpho- logical correlates with smoking. The original series contained 213 volunteers but 46 dropped out for various reasons, and the age range was most extensive from 20 to 59 years of age. Damon's non-smoker category con- sisted of subjects not currently smoking and had never been regular smokers. Cigar and pipe smokers were combined with cigarette smokers, and the statistical analysis was based on the biserial correlation coefficient. As a result of his analysis, Damon found that smoking was associated at the 5 percent level with bi-iliac/biacrominal breadth. subscapular skinfold, ectomorphy, and physical activity; and at the 1 percent level with weight, height/cube root of weight, endomorphy and somatotype group. Smokers of all grades had very similar levels of activity. On the other hand, the most active and the least active men smoked more than those of average activity-a finding which reflects a curvilinear regression of smoking on activity. Damon concludes: "The results show a consistent and significant tendency . . . for lean men to smoke more than stout or fat (but not mus- cular) men . . . higher cholesterol levels among smokers . . . contrary to findings previously reported, smokers in this series were no less masculine in physique, were no more active and consumed no mnre alcohol than non-smokers." PROSPECTIVE STUDIES The most extensive study of morphology as related to smoking habits is Seltzer's prospective study of 922 H arvard alumni 13 years out of college, whose physical characteristics were recorded when they were under- graduates (6) . The investigation was concerned with the morphological characteristics of different classes of non-smokers, cigarette smokers, pipe smokers, and cigar smokers, in a selected male population in order to ascertain the extent to which different smoking classes are phenotypically and genotypi- tally conditioned. The morphological material consisted of a series of anthropometric measurements taken in the fall of 1942 as part of the routine Harvard College medical examination. A total of 12 measurements were obtained of various parts of the body, from which 10 body ratios or indices were computed. When the morphologic data were collected, there was no 385 prior consideration or knowledge of their ultimate use in this correlative study with the subjects' subsequent smoking histories. Information With respect to the smoking habits of these Harvard men was obtained in the fall of 1959 through the medium of a questionnaire (81 percent response). The questionnaire covered approximately 16 years of smoking history and the subjects at the time of completing the questionnaire averaged 35 years of age, a period of maximum lifetime smoking experience. . AS far as smoking categories are concerned, an attempt was made to obtam groupings as nre. cisely differentiated as possible. The primary ClaSSifiCatiOn separated the subjects into non-smokers and smokers. The non-smoker was defined as a person who had never smoked at all or had attempted an occasional smoke during his lifetime. Individuals who smoked occasionally but not every day were excluded from the non-smoker category. The smokers were subdivided into exclusive groupings of cigarette only, cigar only, and pipe only in accordance with the form of tobacco used. All who regularly used more than one form of tobacco were omitted from this particular classification. For the analysis of degree or rate of cigarette smoking, there was a breakdown into five subgroups from occasional to 2f packs a day. The prospective nature of the study, with the availability of the physical measurements made during the college years, had the special advantage of representing a level of morphological status undifferentiated by individual variations resulting from modes of habit, diet, physical activity, health and disease of the subsequent adult years. The analysis was divided into three parts: comparison of non. smokers and smokers, variations among smokers according to form of smok. ing, and variations among smokers as related to degree or rate of smoking. The comparison of 234 non-smokers and 688 smokers showed that the two groups were significantly differentiated both in morphologic dimen- sions and proportions. ln every instance, the smokers had larger mean dimensions than the non-smokers, and in all but one instance these differences were statistically significant. Smokers were consistently greater than non- smokers in height, weight, and in the dimensions of the head, face, shoulders, chest, hip, leg, and hand. Similarly, the smokers of cigarettes only, pipes only, and cigars only had larger mean dimensions than those of the non-smoker category. In addition, in eight out of ten bodily indices or proportions the smoker types showed mean deviations from the non-smoker that were all in the same direction and indicative of the same trend. A consistent graded pattern of differentiation into a specific order of arrangement of non-smokers, cigarette only, pipe only, and cigar only smokers, in that order, was found. Thus, for example, in the case of weight, the cigarette only smokers were 4.37 pounds heavier than the non-smokers, the pipe only smokers 6.59 pounds heavier, and the cigar only smokers 10.41 pounds greater mean body weight. Analysis of the data dealing with amount of cigarette smoking did not show a regular significant body build differentiation according to rate or degree of smoking, but there were suggestions of a positive linear trend from the lightest smoking category to the "1 to 2 packs daily" followed by a downward trend of the maximum "2f packs daily" smokers. Of all the morphological studies, this prospective study appears to present the best data available. Nevertheless, the Harvard students comprise a highly selected sample. 386 CONCLUSION The available evidence suggests the existence of some morphologic differ- ences between smokers and non-smokers, but is too meager to permit a conclusion. REFERENCES 1. Damon, A. Constitution and smoking. Science 134: 339, 1961. 2. Diehl, H. S. The physique of smokers as compared to non-smokers. A study of university freshmen. Minnesota Med 12: 421, 1929. 3. Heath, C. W. Differences between smokers and non-smokers. AMA Arch Intern Med 101: 377, 1958. 4. Parnell, R. W. Smoking and cancer. Lancet 1: 963, 1951. 5. Seltzer, C. C. Masculinity and smoking. Science 130: 1706, 1959. 6. Seltzer, C. C. Morphologic constitution and smoking. JAMA 183: 639,1963. 7. Thomas, C. B. Characteristics of smokers compared with non-smokers in a population of healthy young adults, including observations on family history, blood pressure, heart rate, body weight, cholesterol and certain psychologic traits. Ann Intern Med 53: 697, 1960. 387