CHAPTER 1
Self-image, Medicine, and Drug Use
Kurt W. Back and Deborah A. Sullivan
The problems associated with drug use can range from a physician's attempt to obtain a patient's acceptance of a regime of prescribed medicines to the prevention of frivolous or even dangerous drug abuse. The variation among individuals in their use of drugs depends on both their willingness to be changed and their acceptance of a chemical means of inducing this change. Some types of drugs are more socially acceptable than others. In fact, one could propose a spectrum of drugs which includes, at one end, those prescribed to treat acute infectious diseases or physiological disorders and, at the other end, those prohibited drugs used for "recreational" purposes. Motivation for the use of a drug may depend on the specific situation or the specific effect of the drug to be used. However, a predisposition to use or not to use drugs in any situation also may explain some of the variation in drug use patterns among individuals.
Previous studies (Brehm and Back, 1968; Brehm and Davis, 1971), have shown that the use of drugs is related to general personality tendencies, such as a "great desire to be changed, devaluation of the self-image, belief in the beneficial influence of somatic means, and even a positive reaction and curiosity about the effects of physical agents" (Brehm and Back, 1968). A questionnaire, devised to investigate this area, led to the identification of five factors which represent several different personality traits related to a preference for drug use, particularly the use of drugs with strong psychological effects such as opiates, energizers, and hallucinogens.
These previous studies principally include young subjects who were conscious of stimulating or recreational drug use. Moreover, the drugs listed in the questionnaire were mainly psychotropic, mood-changing drugs, and the setting of the data collection may have encouraged the subjects to interpret "drugs" in this sense. In the current study we focus on middle-aged and elderly subjects and shift the attention more toward drugs used for medicinal purposes, thereby looking at the general conditions of drug use.
Method
The data are derived from a multidisciplinary study of middle-aged and elderly population (Duke Second Longitudinal Study). This research uses data collected in two separate parts of the program. One source is the medical history section of the physical examination, where the respondents were asked to check the amount of drugs and medicines which they had taken in the preceding year. This list includes both prescription and nonprescription drugs and those used for specific physical ailments, as well as those used for general onerous conditions. In addition, the use of alcohol and tobacco was checked. Table A—1 shows the extent of use of the different drugs. They are classified into those which deal with specific ailments (medicines), with nonspecific complaints (medicinal drugs), and with nonmedical social use (social drugs).
Table A—1. Extent of use of different kinds of drugs.
The rest of the data are taken from the social history questionnaire. Besides the thirty-five statements that yielded the drug-use personality factors in the previously mentioned studies, a number of standard personality scales are used. These include the sick-role index (Mechanic and Volkart, 1961), the internal-external scale (I-E scale) (Rotter, 1966), the positive-negative affect indices (Bradburn and Caplovitz, 1965), and a series of semantic differentials. In these semantic differentials the concepts "myself," "how I would like to be," and "how I appear to others" are rated on seven bipolar scales. The Euclidian distance between these concepts (the square root of the means of the squared difference between the scale scores) gives an impression of the discrepancies between the three concepts. The distance between "myself" and "how I would like to be" is termed selfevaluation, that between "myself" and "how I appear to others" is termed persona, and that between "how I wpuld like to be" and "how I appear to others" is termed ideal persona.
Results
Factors
Five factors emerge from the principal components analysis of the thirty-five statements related to sickness and drug use among the middle-aged and elderly sample (table A—2). Items related to a dissatisfaction with oneself (e.g., seeing oneself as too nervous, anxious, self-conscious, or tired) and a desire for some physical action to overcome personal limitations dominate the first factor. As in previous studies, this factor is labeled "insecurity." The second factor accounts for 10.5 percent of the total variance and loads highly on those items related to seeking medical attention and showing concern about health. These items appear to measure the degree of willingness on the part of the respondents to accept a sick role. In contrast, the third factor loads highly on items related to fear of medicine or drugs. Paradoxically, curiousness about the personal effect of a "mind-expanding" drug also loads highly on this factor. The fourth factor also indicates a fear of medicine or drugs. However, the fear embodied in this factor is related more directly to a fear of the loss of personal control that might result from medicine or drugs rather than the medicine or drugs themselves. The final factor deals with curiosity about mind-changing drugs. The five factors together account for 44.2 percent of the total variance.
A comparison of this factor structure with that found in the previous studies of young subjects reveals several striking similarities. The insecurity and sick-role factors are almost identical. The difference in the insecurity factor consists of a high loading on the statement "In the past I faked illness to get out of something I haven't wanted to go do" for the older sample and high loadings on three other statements for the student sample ("I get very afraid if I don't know what is happening to me," "I am more concerned than most people about my bodily feelings," and "I wish that all my problems could be solved by taking a pill").
The only difference in the composition of the sick-role factor is a high loading on the relative concern about personal health statement in the older sample that was not found in the younger sample. However, the higher order of the sick-role factor in the older group, compared to the younger group where it was the third factor, suggests the greater importance of health for older respondents.
The "fear of loss of control" factor was second in importance for the younger sample, although it is the fourth factor in the older sample. This reflects the greater concern with mind-changing drugs in the younger sample. The lesser interest in nonmedicinal aspects of drugs in the older sample is further supported by the smaller number of statements heavily weighted in the fifth, curiosity, factor.
The major difference in the factor structure between the two samples is the appearance of a factor suggesting a general fear of medicine or drugs in the older group. In contrast, two of the statements in this factor ("I would avoid taking drugs when I'm sick for as long as I could" and "In general, I tend to avoid taking medicine and drugs") loaded with two of the statements from the "fear of loss of control" factor ("I might do something under the influence of drugs that I would not normally do" and "In general, I tend to avoid taking medicine and drugs"), forming a factor that was labeled as "denial of effects" for the younger sample. This "denial of effects" factor does not emerge in the older sample.
Drug Use
Table A—3 shows the relationship between the five personality factors and the indicators of drug and medicine use. For women, use of medicine is significantly correlated with three of the personality factors: insecurity, sick role, and fear of medicine. For men, none of the personality factors is significantly correlated with use of medicine, although acceptance of a sick role is just below the significance-level cutoff. This may indicate that the use of specific and usually prescribed medicines relates to an acceptance of a clearly defined sick role and that this definition is more salient among women than men. The importance of the definition of the sick role may partially explain the curious positive correlation factor 3 (fear of medicine) with taking medicine among women: the fear of the possible effects of medicines may be just the impulse which makes women consult a physician and, therefore, have medicines prescribed. This is further supported by the relationship between insecurity and the use of medicine among women.
It is interesting that there is no significant relationship between insecurity and use of medicines for men, although insecurity is strongly related to use of medical drugs (many of which do not require a prescription) for both sexes. This suggests that insecurity in general is related to a desire for change and an acceptance of a chemical means of inducing change. However, men appear more reluctant than women to seek a physician's aid.
None of the personality factors was significantly associated with the social drugs tobacco and alcohol; however, the negative correlation with sick role for women and the positive correlation with fear of medicine for men are just below the significance level. This might suggest tentatively viewing tobacco and alcohol, especially the latter, as an alternative to acceptance of the sick role, but this relation is different in the two sexes, corresponding to their differences in the willingness to accept a sick role and consult a physician.
Table A—3. Correlations of attitude factors with type of drug use by sex.
Further tests of the relationships are provided by multiple regression analyses of the different kinds of drug consumption. Several other variables which measure similar dimensions of self-acceptance, some of which have been shown to be important in previous drug studies (Brehm and Back, 1968; Brehm and Davis, 1971), are added to the five factors as predictor variables. These include the I-E scale, the positive-negative affect indices and the self-evaluation, persona, and ideal persona concepts formed from semantic differentials of "myself," "how I would like to be," and "how I appear to others."
Table A—4 shows the results for men and women. The predictor variables are added in a step-wise program down to an F-level of 1.00. Variables with lower F values are eliminated; their inclusion would not change the multiple correlation to any appreciable degree.
The multiple correlations are consistently higher for women than for men and highest for medical drugs, followed by medicines and social drugs. This, then, is the rank order of the predictive capacity of our personality variables for the different dependent variables.
The regression analysis adds further insights to the influence of personality traits on drug use. Use of medical drugs in both sexes depends on general unhappiness-insecurity, negative feelings, and the difference between how one appears to others and how one would like to be (ideal persona). These strong relationships are the most important results of the study.
For medicines, sex differences are important. Men are influenced mainly by ideal persona and positive affect (i.e., having had positive feeling during the preceding week). Several other variables have only a minor effect. Women, however, are influenced to take medicines mainly by the fear of medicines, acceptance of sick role, insecurity, and fear of loss of control, as well as by the difference between how they are and how they appear to others (persona). A l though there is some similarity between men and women, men seem to have a more positive approach to medicines than women have. Women seem to be influenced mainly by fear.
Table A—4. Regression analysis of medicine-drug use.
A similar sex difference occurs in the prediction of the use of the social drugs tobacco and alcohol. For men, fear of medicine is important; for women, a variety of variables show some effect, although none is significant. This finding provides further support for the contention that men who fear medicines or mental drugs may be using alcohol as a substitute.
Discussion
The results of this study lend some support to the hypothesis that personal attitudes and disposition are important factors in the use of medicines and medical drugs as well as of mood-changing drugs. Measures which had been shown to be related to mind-expanding drug use among young adults in a previous study are also related to the use of drugs for medicinal purposes in an elderly population both to treat a specific psychological disorder and to treat nonspecific symptoms.
While similarities in predispositions toward drug use among young and old justify this kind of approach, we must also note the differences. The original scale factored in a slightly different way, leading to clusters of items concerned with illness and the use of medicine and not with the search for stimulation and curiosity. Further, the clearest pattern was found with the medical drugs; medicines seem to be more closely connected with the demands of the illness, and the social drugs alcohol and tobacco may be determined by social influences.
The middle-aged and elderly respondents try to obtain a balance between their general anxiety and mood, their feelings about illness and medicine, the way they see themselves, their impression on others, and how they would like to be. These personal preferences lead to consumption of drugs, depending on the availability and use of medical and pharmaceutical services. The different patterns exhibited indicated the alternatives that are possible for accepting a rational relationship between medical care and the reasons, fears, or alternative attractions which might interfere with a purely rational use of health care. Thus, the data give an indication of the social-psychological determinants of drug use.
