VoL I No. 4 Fall 1986
Journal of Pain and Symptom Management
What is Behavioral About Behavioral Medicine?
Handbook of Behavioral Medicine Edited by W.D. Gentry Published by Guilford Press New York, 1984, 575 pp, $40
Reviewed by Richard E Sloan A p r o m i n e n t physiological psychologist remarked to me in 1978 that a then recent study by Levine and Fields had finally ended the mind-body problem. He reasoned that since the study demonstrated that placebo analgesia, like opiate analgesia, could be reversed by the opiate antagonist naloxone, it was clear that physical pharmacological agents caused psychological phenomena. I remarked in response that I saw no resolution to the problem; what these findings showed to me was simply that the focus of the problem had shifted from the effects of psychological p h e n o m e n a on physical ones to the impact of physical agents on psychological phenomena. In no way did the study explain how a psychological factor, the placebo effect, could produce physiological consequences which could in turn be reversed by ~/drug. Many )'ears later dimensions of the same great problem remain unresolved and hidden in a new arena: that of behavioral medicine. The Handbook of Behavioral Medicine, while containing excellent chapters on a variety of issues in the field, raises substantive questions about what constitutes the field, how we develop the-
Richard P. Sloan, PhD, is Coordinator of Behavioral Medicine at Columbia-Presbyterian Medical Center, New York, and Associate Professor of Psychology at the State University of New York, College at New Pahz.
ories to account for its findings, and what assumptions underlie these theories. Tile questions relate as much to the issue of mind and body as they do to the n a m e of the field: behavioral medicine. In tile name of behavior we find a wide variety of phenomena: sociocultural factors (Syme), personality (Krantz and Glass), social and psychological resources (Gentry and Kobasa), coping and adaptation (Lazarus and Folkman), and decision making (]anis). Mentioning these topics is not meant to be critical of the chapters; they are excellent, representing the views of leaders in behavioral medicine. These remarks are addressed, rather, at the views of the field, represented in this case by the editorial judgments of what constitutes the field. Substantive differences exist between behaviors and such things as beliefs, attitudes, personality, and cognitions. Subsuming all of these under the rubric o f "behavior" conceals important differences, not only in the nature of these events but also in the way we construct theories to relate them to health outcomes. As Gentry points out in his introductory chapter, different views of the field exist. One may take a "hard" view of behavioral medicine and argue that it should restrict itself to actual behaviors and their relationships to matters of health. Such a view might be represented by some of the work described by Ader and Cohen, in their discussion of the conditioning of the immune response. A broader view of tile field might instead understand behavioral medicine to be just about anything medical which is excluded by conventional medicine, for example, how patients make medical decisions, how personality influences physical health, what sociocultural factors relate to disease etiology, etc. That is, one may study health psychology or even
VoL I No. 4 Fall 1986
biopsychosocial medicine. Plainly, the inclusion of these and other related topics in the Handbook suggests that Gentry views the field in this broad way. In fact, there is good reason to believe that this is the d o m i n a n t view of the field. I would like to argue that this view results in conceptual confusions which will have an impact upon the future of the field. This misconception is rooted in the failure to distinguish behaviors from nonbehaviors and ultim a t e l y relates to o u r views o f the still unresolved mind-body problem. Two issues arise: 1) what is behavioral about behavioral medicine? and 2) how are behaviors and health outcomes related? In regard to the first issue, behaviors are, or should be, defined as movements of the body. Such things as eye blinks or hand waving are behaviors. So is smoking a cigarette. By virtue of their "objectifiable" nature, behaviors can be measured. The program of John Watson, who in 1913 launched the behaviorist assault against the vagaries of introspectionist psycholog3; sought this precision. By restricting psychology to the study of behavior, Watson thought that he was simuita. neously elevating psychology to a level of precision rivaling physics. The relevance of this issue for behavioral medicine arises when one considers the problems Watson immediately encountered when he tried to apply his views to the real world. Having learned, in about 1915, of the stimulusresponse psychology developed by Pavlox; Watson (1930) explains how the stimulus of food produces the response of secretion of saliva. Both the stimulus and the response can be measured with great precision and without reliance upon vague introspective data. Behaviorism promised the precision of physics to a psychology dependent upon the "subjective" data of introspection. So it might come as a surprise that later in the very same chapter ("How to study human behavior"), Watson provides us with the following examples of applications of b e h a v i o r a l t h e o r y : he asks a b o u t the "responses" to the "stimuli" of "overthrow of monarchy; formation of soviet government; war; prohibition; easy divorce; elimination of hereditary wealth" (page 43). Similarl); he asks us to determine what "stimuli" might lead to the "responses" of "marriage under modern financial pressure; truthfulness; continence in great cities where social control is difficult" (page 43). As these examples indicate, the terms
Journal of Pain and Symptom Management
"stimulus" and "response" lose all meaning under Watson. Virtually anything can be a behavior. As the terms lose their meaning, the power of behaviorism to explain them erodes. The problems in defining the subject matter of behavioral psychologies remain to this day, even though Watson and his psychology are long gone. Operant psychologists and social learning theorists no less than those who study classical c o n d i t i o n i n g systemically fail to address the problem in any satisfactory way. Either they remain close to their theoretical assertions and study behavior that is precisely measurable and of precious little interest or, tugged by the need to develop a psychology that addresses real-life phenomena, abandon their interest in precision and talk about applied topics in the same wa); save for tile jargon, as would any non-psychologist. This lack of precision in defining its principle subject matter leads behaviorism, and behavioral medicine for that matter, to the second issue, that of the theoretical connection between the things to be explained and the putative causal agents. Watson believed that by creating a completely "objective" psychology, he had eliminated once and for all the mindbody problem. Issues of mind and body also plague behavioral medicine. Taking the term "behavior" literally leads to two advantages: 1) the precision that comes with defining behaviors as measurable movements and 2) the conceptual clarity o f having both the cause, behavior, and the effect, health outcomes, within the same domain: the physical world. The construction of theories to relate behaviors to health outcomes is very much different from the construction of theories to relate such insubstantial things as emotions, feelings, attitudds, ahd beliefs, to health outcomes. Taking this narrow vie~; however, causes the field to become arid, since the topics of real interest, for example, loneliness in its relationship to immunosuppression, are beyond the limits of that domain. To take the wider view of the field and go beyond behavior has the advantage of including such things as cognitions, attitudes, feelings, etc. However, this approach brings with it a different problem: that of interactionism, the mutual influence of physical and psychological events, without us having any idea how to account for it. This conflict between precision and relevance, one which has concerned psy-
1hi. 1 No. 4 I+d11986
chology since its inception, is still with us. But as William James wrote, the boundaries o f mental life are vagne. Better that its science be simi. larly vague. At the heart o f the issue, then, is the question o f how behavioral medicine views itself. If, in fact, researchers are interested in the effects o f beliefs and feelings as well as behaviors u p o n health outcomes, then the)' must construct theories which relate these presumed causal factors to the resultant health conditions. This requires theory which recognizes the natures o f the related constructs. But since we have no such theories, we are left only with the possibilit}' o f u n c o v e r i n g statistical r e l a t i o n s h i p s between incommensurate events, in much the same way that the Babylonians, having successfull}' developed equations to predict the movement o f heavenly objects, attempted to apply their predictive equations to terrestrial events such as earthquakes and plagues o f locusts (Toulmin, 1961). Such application underscores the failure of Babylonian science in this regard: it did not understand that earthquakes and plagues o f locusts were different kinds o f events f r o m the m o v e m e n t o f planets and stars. D e m o n s t r a t i n g statistical relations between events must be regarded as a first step in the search for theories to relate these events. But theories relating psychological events to health outcomes must not assume that the principles which apply to one sphere automatically apply to the other. In what sense, then, is the Handbook of Behavioral Medicine about behavioral medicine? T h e answer is "not much" as examination o f most chapters will reveal. Herd's wonderful and encyclopedic chapter on cardiovascular disease a n d hypertension, for example, is almost exclusively about physiology. Occasional references to behavior are made but careful examination reveals that the "behaviors" are often emotions, stress produced by a mental arithmetic exercise, anxiet}; or personality factors. Syme's chapter on sociocultural factors and disease etiology is also about things o t h e r than behaviors. These topics, to be sure, are worthy of investigation in their relationships to health outcomes but they are not behaviors. If we are intent to study them in their relation to health, we must have theories which recognize their nature. One chapter, that written by Krantz and Glass, approaches some o f the issues discussed above, hut does not address directly the issue o f
Journal of Pain and Symptom Management
behavior. It does, however, address the issue o f the association o f personality styles and other psychological factors, eg, helplessness and disease. While the chapter nicely reviews theories regarding the relationships between disease outcomes and psychological factors, the theories do not adequately address what mechanisms might account for these relationships. This remains true even in the section devoted to the psychophysiological mechanisms presumably underlying coronary heart disease. T h e statistical association between sympathetic nervous system hyperreactivity and heart disease presents an interesting problem, one in which both o f the events to be related belong to the same domain. However, how do we develop a t h e o r y to a c c o u n t for the r e l a t i o n s h i p between feelings o f control and SNS hyperreactivit}; which belong to completely different domains? This analysis is not meant to diminish, in an)' wa)~ this important work; it is meant only to suggest that behavioral medicine must not be satisfied with m e r e statistical associations between events in the place of theory. So where does this leave behavioral medicine? Hopefully, with an identity crisis, something characteristic o f a new field in science. One can hope that as the field matures, the crisis will resolve itself and the field will recognize the challenges before it.