Home Curriculum Vitae Publications Conference Reports Forthcoming Extramural Colloquia Expert Testimony Teaching Healthcare The Human Ecology of Memory Research Archive Publications and Reports Rants


Somatization as Illness Behavior

John F. Kihlstrom and Lucy Canter Kihlstrom

University of California, Berkeley


Institute for the Study

of Healthcare Organizations & Transactions


Note: an edited version of this article appeared as Kihlstrom, J.F., & Canter Kihlstrom, L.  (2001).  Somatization as illness behavior.  Advances in Mind-Body Medicine, 17, 240-243.

Somatization disorder, in which a patient complains of physical symptoms for which there are no demonstrable organic findings or known physiological mechanisms, has been characterized as "medicineís unsolved problem" (Lipowski, 1987, p. 294). Certainly it is a problem: the per capita expenditure for health care of somatizing patients is up to nine times the average (Smith, Monson, & Ray, 1986), including repeated consultations with primary-care physicians, unnecessary hospitalizations, and loss of work time. They resist referrals to mental health practitioners, and also appear to use complementary and alternative medicine (CAM) at a relatively high rate -- perhaps because of more frequent consultations and improved provider-patient relationships offered by CAM practitioners (Garcia-Campayo & Sanz-Carrillo, 2000).

Although the NIMH Epidemiologic Catchment Area study reported a rate for somatization disorder of only 0.1% in the general population (Robins et al., 1984), the syndromeís prevalence is much higher among patients seen primary-care clinics, specialty clinics, and psychiatric consultation/liaison services (e.g., Kellner, 1990). Although generally considered a disorder of middle-aged women, a recent survey found that almost 11% of undergraduates at a major American public university, including almost 15% of women and almost 7% of men, reported enough somatic complaints to cross the DSM-III-R threshold for somatization disorder (Canter Kihlstrom, Marsh, & Kihlstrom, 1998; see also Kihlstrom & Canter Kihlstrom, 1999).

Somatization disorder, like conversion disorder and other "somatoform" disorders recognized by DSM-IV, has often been construed as an illustration of the "mind-body" interaction -- as the translation of emotional distress into somatic symptoms, in contrast with psychologization (e.g., Kirmayer & Robbins, 1991). Perhaps the increased interest in complementary, alternative, and integrative medicine will revive medical interest in psychosomatic medicine and other aspects of mind-body interaction. On the other hand, the psychosomatic disorders are currently in bad repute. Classic psychosomatic syndromes like ulcers are now attributed wholly to physiological processes that have nothing to do with emotion (e.g., Hyman, 1994; but see Overmier & Murison, 1997). Both DSM-I, published in 1952, and DSM-II (1968) contained a major category of psychophysiological autonomic and visceral disorders, further classified according to the organ system involved. In DSM-III (1980) and DSM-IIIR (1989), the psychosomatic and psychophysiologic disorders were listed as "psychological disorders affecting physical condition". But DSM-IV contains no such category: it is open to "mental disorders due to a general medical condition" (Code 293.83), but "psychological factors affecting medical condition" are relegated to the back of the book. "Psychosomatic" has gone the way of "neurosis" and "psychosis".

McWhinney and his colleagues seem to applaud this trend in their target article. In their view, the very notion of psychosomatic illness reflects an outmoded Cartesian dualism between mind and body, a stance that has been rendered obsolete by the advances of modern medicine and neuroscience (McWhinney, Epstein, & Freeman, 1997). If the mind is what the brain does, and the brain controls the body, then there is nothing "abnormal" about the translation of mental states into physical states: emotion is embodied, along with cognition and motivation, and thatís all there is to it. Or, at least, thereís no point in trying to figure out how the mind affects the body, because mind and body are the same thing.

Such a rejection of psychosomatic concepts is in line with reductionist trends within psychology, but it may be a mistake to include somatization disorder in its sweep. This is because somatization does not involve somatic symptoms that can be confirmed and explained by physical examination or laboratory test, but rather somatic complaints. The patient may complain of cardiovascular or gastrointestinal symptoms, but physical examination and laboratory testing reveals no evidence of tachycardia, or gastric lesions. In this respect, somatization disorder is similar to body dysmorphic disorder and hypochondriasis, some factitious and eating disorders, and malingering. All represent "illness without disease", in the words of the Harvard Mental Health Letter (Illness without disease -- Part I (somatoform disorders), 1999; Illness without disease -- Part II (somatoform disorders), 1999) and it is the inability of the health-care provider to confirm the patientís physical complaints that makes them so frustrating for all concerned. By contrast, in the conversion disorders, and in what used to be called the psychosomatic and psychophysiological disorders, the physician can confirm the patientís complaints -- even if their causes remain mysterious.

Put another way: while the psychosomatic syndromes may be properly construed as physical illnesses with psychosocial causes, it seems likely that somatization is better construed as illness behavior (Mechanic, 1962) -- behavior which, like all behavior, must be understood in terms of the patientís personal experiences and life circumstances. The somatizing patient may be anxious or sad, angry or resentful, unhappy in marriage, or frustrated at school or work, or have any of a host of problems in living. This is what the complaints are really about, not the heart or the gut, and these are the problems that have to be addressed. Nothing is embodied in somatization at all. Behavioral neuroscience, psychoneuroendocrinology, and psychoneuroimmunology may help us understand the psychosomatic disorders, but they cannot help us to understand somatization, for the simple reason that the body is not involved in the phenomenon except in the trivial sense that the body is involved in all behavior. The symptomatic complaints of the somatizing patient may well be bodily expressions of emotional distress, and attempts to use oneís body as a vehicle for social control, but that is not the same thing as saying that they are physical embodiments of that distress. In some sense, somatization is an aspect of personal identity: somatizing patients identify themselves as a sick people, just as patients with body dysmorphic disorder identify themselves as ugly people (Kihlstrom & Canter Kihlstrom, 1999), and behave accordingly.

To propose that somatization can be understood only by examining the person in social context (a point on which we agree with MacWhinney et al.) is not to revive an old-fashioned psychodynamic view that symptoms must be interpreted to reveal the patientís unconscious conflicts. In fact, much of the basis for somatization probably lies outside individual patients, in the social context in which their behavior takes place. Just as cultural factors affect the individualís experience and presentation of physical symptoms, so some cultures may prefer somatization over psychologization as the means for the individualís behavioral expression of distress. In fact, some medical anthropologists have argued that psychologization is something of a Western invention, and somatization more the norm in the rest of the world (e.g., Kirmayer, 1984; Kleinman & Kleinman, 1985). If somatization is to some extent a culture-specific disorder, abnormal only in the industrialized West, there is no point in adding bio to psychosocial: cultures may differ, but bodies are everywhere alike.

Even within a culture, social factors, from the interpersonal to the institutional, may affect the degree to which somatization occurs. Because mental patients remain stigmatized to a considerable a degree in our society, for example, the benefits of the sick role accrue to physical rather than mental illness. Someone with heart palpitations can be excused from attending a family function, but someone with lingering resentment will be expected to overcome it for purposes of maintaining harmony. In addition, while somatizers are generally considered to be "difficult" patients, difficult patients can also be labeled as somatizers. A recent study of medical practice found that physicians were more likely to diagnose symptoms as medically unexplained (the core of somatization) if they perceived their interactions with the patient as negative rather than positive (Nimnuan, Hotopf, & Wessely, 2000). Because medically unexplained symptoms lie at the core of somatization, to some extent somatization may reside in the physician, not the patient.

In addition, changing diagnostic standards may affect how the patient is labeled. In DSM-III and DSM-IIIR, somatization disorder was considered if the patient presented a high number of symptomatic complaints, without restriction on their distribution. However, DSM-IV requires that at least one of these symptoms come from the sexual and reproductive sphere. Such complaints come almost wholly from women, and in fact some of the relevant symptoms, such as irregular menses and vomiting throughout pregnancy, can only be reported by women. In a study of symptomatic complaints among college students (Canter Kihlstrom et al., 1998), application of the DSM-IV criterion for somatization not only reduced the percentage of subjects crossing the threshold for somatization, from almost 11% to just over 8%; but it also greatly increased the ratio of women to men, from 2:1 to 7:1. In other words, DSM-IV effectively redefined somatization as a "female" disorder. Boys and men with multiple unexplained medical symptoms are now less likely than before to be classified as somatizers.

The very structure of the healthcare system may be an important factor in somatization. Medical procedures for diagnosis and treatment naturally focus on anatomy and physiology, and may encourage somatic rather than psychosocial attributions for distress. Attempts to manage somatization within primary care, by performing regular physical examinations and foregoing special diagnostic tests or hospitalization, may control costs and improve patient satisfaction over the short run (Morriss et al., 1999; Smith, Monson, & Ray, 1986), but may fail over the longer term (Kashner, Rost, Smith, & Lewis, 1992). Referral to a psychiatrist or other mental-health professional fails because it clearly communicates to patients that their problems are in their minds, or their social relationships, rather than in their bodies -- which is probably true, but not what they want to hear.

In fact, medical specialization itself may play an important role in supporting somatization. A patient with presenting complaints that cannot be verified or explained by an internist is likely to be referred to a specialist, who will spend even more time, and perform even more expensive tests, trying to find something wrong with the patientís body. The availability of such specialists in a modern health-care system provides a ready escape route for both patients and general practitioners, and may effectively delay both the recognition of somatization and its treatment.

Specialization may even play a role in creating new forms of somatization disorder. The classic presentation of somatization disorder, patterned on Briquetís syndrome (Mai & Merskey, 1981), is a patient with multiple unexplained symptoms. However, somatizing patients can also present single symptoms, or multiple symptoms within only a single bodily system. In fact, it has been argued that the tendency of specialists to focus only on that part of the body that lies within their expertise has led to the proliferation of functional somatic syndromes within each medical specialty: irritable bowel syndrome for gastroenterologists, fibromyalgia for rheumatologists, tension headache for neurologists, multiple chemical sensitivity for allergists, and so on. Nevertheless, there appears to be substantial overlap among these functional somatic syndromes in terms of diagnostic features, response to (psychosocial) treatment, and other characteristics, suggesting that they are more alike than different (Wessely & Nimnuan, 1999).

We are not suggesting that somatization, in whatever form it takes, is "all in the mind" of either patients or physicians. Some patients really do suffer from irritable bowels or tension headaches (and, in some cases, the patientís problems may be genuinely "psychosomatic" in nature). These symptoms are a source of discomfort to them and an object of frustration for their caregivers; but they also provide patients with entreí into the sick role and its benefits. However, we are suggesting that somatization may be the wrong place to look for a resolution to the mind-body problem, whether by reinforcing Cartesian interactive dualism, in which the body responds to emotional states, or abandoning dualism in favor of a unified conception of psyche and soma in which emotion is embodied in bodily processes. This is because at least some somatizing patientsí problems do not lie anywhere in their bodies. Rather, they are using their bodies, the language and culture of medicine, and the institutions and processes of the health-care system to express and manage their personal and interpersonal difficulties in a way that would be otherwise difficult or impossible. Somatization may, for some individuals, be an acceptable way of interacting with others in a medicalized world. From this point of view, understanding somatization requires not that we look into the patientís body, but rather into the patientís life and the world in which he or she lives.


Author Notes

The point of view expressed in this paper is based on research supported in part by Grant MH-35856 from the National Institute of Mental Health. E-mail John F. Kihlstrom at kihlstrm@socrates.berkeley.edu or Lucy Canter Kihlstrom at lucyck@uclink4.berkeley.edu. Further information on somatization and other aspects of health and illness behavior is available at www.institute-shot.com.




Canter Kihlstrom, L., Marsh, K. F., & Kihlstrom, J. F. (1998). Multiple somatic symptoms among young adults. Available: www.institute-shot.com/illness_behavior.htm.

Garcia-Campayo, J., & Sanz-Carrillo, C. (2000). The use of alternative medicines by somatoform disorder patients in Spain. British Journal of Medical Practice, 50, 487-488.

Hyman, S. E. (1994). Another one bites the dust: An infectious origin for peptic ulcers. Harvard Review of Psychiatry, 1, 294-295.

Illness without disease -- Part I (somatoform disorders). (1999). Harvard Mental Health Letter, 16(3), 1-4.

Illness without disease -- Part II (somatoform disorders). (1999). Harvard Mental Health Letter, 16(4), 1-4.

Kashner, T. M., Rost, K. M., Smith, G. R., & Lewis, S. (1992). The impact of a psychiatric consultation letter on the expenditures and outcomes of care for patients with somatization disorder. Medical Care, 30, 811-821.

Kellner, R. (1990). Somatization: Theories and research. Journal of Nervous & Mental Disease, 178, 150-160.

Kihlstrom, J. F., & Canter Kihlstrom, L. (1999). Self, sickness, somatization, and systems of care. In R. J. Contrada & R. D. Ashmore (Eds.), Self, social identity, and physical health: interdisciplinary explorations (Vol. 2). New York: Oxford University Press.

Kirmayer, L. J. (1984). Culture, affect, and somatization: I & II. Transcultural psychiatric research Review, 21, 159-188, 237-262.

Kirmayer, L. J., & Robbins, J. M. (1991). Current concepts of somatization: Research and clinical perspectives. Washington, D.C.: American Psychiatric Press.

Kleinman, A., & Kleinman, J. (1985). Somatization: The interconnections among culture, depressive experiences, and the meanings of pain. A study in Chinese society. In A. Kleinman & B. Good (Eds.), Culture and depression (pp. 132-167). Berkeley, Ca.: University of California Press.

Mai, F. M., & Merskey, H. (1981). Briquet's concept of hysteria: an historical perspective. Canadian Journal of Psychiatry, 26, 57-63.

McWhinney, I.R., Epstein, R.M., & Freeman, T.R. (1997). Rethinking somatization. Annals of Internal Medicine, 126, 747-750. Reprinted in Advances in Mind-Body Medicine, in press.

Mechanic, D. (1962). The concept of illness behavior. Journal of Chronic Diseases, 15, 189-194.

Morriss, R. K., Gask, L., Ronalds, C., Downes-Grainger, E., Thompson, H., & Goldberg, D. P. (1999). Clinical and patient satisfaction outcomes of a new treatment for somatized mental disorder taught to general practitioners. British Journal of General Practice, 49, 263-267.

Nimnuan, C., Hotopf, M., & Wessely, S. (2000). Medically unexplained symptoms: How often and why are they missed? Quarterly Journal of Medicine, 93(1), 21-28.

Overmier, J.B., & Murison, R. (1997). Animal models reveal the "psych" in the psychosomatics of peptic ulcers. Current Directions in Psychological Science, 6, 180-184.

Robins, L. N., Helzer, J. E., Weissman, M. M., Orvaschel, H., Gruenberg, E., Burke, J. D., & Regier, D. A. (1984). Lifetime prevalence of specific psychiatric disorders in three sites. Archives of General Psychiatry, 41, 949-958.

Smith, G. R., Monson, R. A., & Ray, D. C. (1986). Psychiatric consultation in somatization disorder. New England Journal of Medicine, 314, 1407-1413.

Wessely, S., & Nimnuan, C. S., M. (1999). Functional somatic syndromes: One or many? Lancet, 354, 936-939.