In a now classic study, Meyer and Haggerty observed 100 persons in 16 families for a year, studying factors related to streptococcus infection. Systematic throat cultures for beta-hemolytic streptococci were taken, as were periodic measurements of antistreptolysin O titer. No relationship was noted between either the number or type of streptococci and either carrier rate or incidence of pharyngitis. However, carrier rate and symptomatology were related to age, season, closeness of contact with an infected person, acute or chronic family stress, and penicillin treatment. ' Most germane to this discussion was the association of acute and chronic family stress to the clinical status of the patients. The authors emphasized the importance of thinking in terms of the "multiple causation" of disease. Although an infectious disease may be identified with a particular pathogen, a variety of factors - including psychosocial environment - can alter the delicate balance between parasite and host.
Starfield and her colleagues characterized this multifactorial approach in a different way in their recent study of health services utilization by children.2 After finding that high users of services were likely to experience a wide variety of health problems, they concluded that an understanding of health service utilization "requires an understanding of the interrelationships of illnesses within individuals rather than a focus on particular illnesses ..."
Taken together, these two studies illustrate the essential principle of psychosomatic medicine - that the onset or severity of disease can be attributed to the combined and interactive effects of the psychosocial and physical environments to which an individual is exposed, as modified by genetic predisposition and prior experience (Figure).3 Psychosocial factors may play greater roles in some diseases, individuals, and circumstances than in others, but the available data indicate clearly that their potential role in any disease process should at least be considered.
The field of psychosomatic medicine has traditionally concerned itself with the impact of psychosocial factors on the onset and incidence of illness. Much of the research in this area has concentrated on disorders such as peptic ulcers, ulcerative colitis, essential hypertension, bronchial asthma, and neoplastic disease. More recently, there have been demonstrations of such influences on a wide variety of infectious and other immune-related diseases as well.3,4 A smaller but equally important body of knowledge demonstrates that a patient's recovery from disease or reaction to medical procedures may also be dependent on psychosocial factors. For example, Greenfield reported that recovery from infectious mononucleosis was related to ego strength. 5 Melamed and others demonstrated that children's responses to surgical procedures could be improved under certain conditions by appropriately preparing them for the procedures, thus reducing both psychological and physiological indices of anxiety.6
THE IMPORTANCE OF STUDYING BIOLOGICAL MECHANISMS
It is possible, of course, to take a strictly "black box" approach to such phenomena, manipulating or observing psychosocial factors on one hand, and observing adaptive or maladaptive outcomes on the other.3·7 However, these effects do not occur in some "mystic" fashion, and a full appreciation of these phenomena ultimately depends on our understanding of the entire chain of events between etiological factors and medical consequences (Figure).
The notion that there is a relationship between "stress" and illness has become an integral part of popular medical culture over the past few years. Many people now believe that the psychosocial environment determines one's susceptibility to disease. Yet, as professionals, we are somewhat at a loss to explain how these effects actually occur, or to provide clinical interventions that might prevent or counteract such effects.
Even if one is not concerned with disease, psychosocial issues may be of importance in the evaluation of clinical findings. Life events may influence such measures as blood pressure, catecholamines or steroids that may be taken for diagnostic or research purposes.
BIOLOGICAL MECHANISMSAN OVERVIEW
Over the years, the primary emphasis in research on biomedical mechanisms of psychosomatic phenomena has been on the adrenocortical system, owing to the pioneering work of Selye and its influence on concepts of stress.8 Selye's research on the General Adaptation Syndrome linked environmental stressors to a nonspecific adrenocortical response. However, it has also been shown that psychosocial factors can cause alterations in many hormones and neurotransmitters, for example, growth hormone, insulin, testosterone, epinephrine, and norepinephrine.3,9,10
Despite the limitations of looking primarily at the adrenocortical system, the volume of work done in this area, especially in experimental animal studies, has amply demonstrated the subtle and specific ways in which such a system can respond to psychosocial stimulation. The adrenocortical response to mild stimulation in the rat depends on such factors as age of the animals, prior experience with similar psychosocial stimulation, whether they are housed alone or with others, the amount of time elapsed between stimulation and blood sampling, and the point in the 24-hour adrenocortical rhythm at which blood sampling is done.3·11 These extensive investigations of adrenocortical response to psychosocial stimulation have, at the very least, alerted us to some of the considerations that should be made in studying the sensitivity and responsivity of other possible mechanisms in both human and animal studies.
The development of micromethods has made it possible to simultaneously examine a number of potential mechanisms under the same conditions of stimulation. Such investigations have underscored both the complexity and specificity of these phenomena; some measures may increase under a given stimulus situation, while others decrease or remain unchanged. Response patterns may change as subjects become adapted to a stimulus situation or learn to cope with it. For example, Monjan and Collector reported that auditory stimulation of the rat either enhanced or suppressed immunoreactivity, depending on the length of exposure to the stimulation. 12 Mason and his colleagues found that hormone responses of monkeys engaged in operant conditioning procedures depended on whether the animals were in a learning or retention phase of the procedure; some measures responded only to the learning phase, whereas others responded to stimulation in the same manner during both phases of the procedure. 10 Some authors, in reviewing this work, have emphasized the importance of looking at patterns of response if the nature of biochemical mechanisms is to be realistically appraised.9,10
The last 10 years have seen a growing attention to immune mechanisms of psychosocial influences on disease. In a review of research in human subjects, Palmblad reported relationships between immune responses and bereavement, hypnosis, ego strength, life events, vulnerability, and schizophrenia.13 A recent study of medical students demonstrated that life events, loneliness and anticipated examinations were related to changes in natural killer cell activity.14
Indices of immune response in these studies have typically involved a number of isolated measures of immune function, such as neutrophil phagocytosis, IgA, antibody titers, and lymphocyte cytotoxicity. In addition, relationships between these measures and psychosocial factors have been largely correlative, and thus suggest possible mechanisms.7 Thus, although a given psychosocial experience may alter a measure of immunity, this change may not necessarily reflect a change in resistance in a clinical sense. Future research will have to examine the influence of psychosocial factors on regulation of the immune system, taking into account both the complexity of the system and the possibility that a given psychosocial state or event may either suppress or enhance immunologic activity.7·15 It can at least be said at this point that the immune system, perhaps in interaction with the neuroendocrine system, is responsive to a variety of psychosocial situations. The extent to which these effects have adaptive value in terms of mediating either increased or decreased resistance to disease remains to be demonstrated in a more systematic way.
A fuller understanding of the ways in which psychosocial factors alter resistance to disease will depend on further investigation of the biological mechanisms of these effects. Previous work has been limited to some extent by the state of biochemical technology and by the frequent adherence of investigators to conceptual frameworks that were excessively narrow in scope. Also, as noted above for work on the immune system, most of the research on biological mechanisms, even in animals, is only circumstantial, in the sense that relationships between mechanism and disease are, at best, correlative rather than causal. Furthermore, although psychosocial stimulation has often been shown to cause consistent changes in physiologic responsivity, these changes may or may not be of adaptive or clinical significance.
The more comprehensive approach to this work in recent years has been accompanied by an increased acceptance by researchers and practitioners in "traditional" medicine. The latter change is vital as it has paved the way for more fruitful collaborative relationships that have done much to enhance our understanding of these mechanisms. Additional prospective and experimental studies, both in animals and humans, will help in establishing causal chains between psychosocial factors, hypothesized mechanisms and resistance to disease. A concurrent appreciation for the complexity of interaction among the neurological, endocrine and immune systems and their respective components in their roles as mechanisms will necessitate a multifactorial approach to these relationships, that is, the simultaneous observation of a system of hypothesized mechanisms and their interrelationships, utilizing appropriate statistical procedures.7,10,13
It must also be pointed out that there has been little, if any, systematic research on the role of such mechanisms in children. Thus, little is known about whether the impact of psychosocial factors on disease is basically different in children than in adults, whether the mechanisms of such effects are similar, or whether psychosocially induced effects on the endocrine or immune systems early in life have lasting developmental effects. Such investigations should greatly enhance our understanding of psychosomatic phenomena in both children and adults.
GUIDELINES FOR THE PRACTITIONER
A number of general points that are relevant to clinical practice can be made based on the current state of knowledge about the relationship between psychosocial factors and disease:
1) Children who are frequently ill tend to have a variety of diseases and problems.2 Psychosocial factors are especially likely to play a role in the illnesses of such children. "Many times," writes Rosch, "it is more important to know what kind of patient has the disease than what kind of disease the patient has."16
2) Disease should not be thought of as either organic or "psychosomatic." It is best to assume that psychosocial factors may play a role in any disease in interaction with other factors in the environment or history of the patient. 17
3) Psychosocial factors can either increase or decrease a patient's resistance to disease, depending on a number of factors.3 Therefore, one should not assume that specific psychosocial influences will always be detrimental to the organism.
4) A patient's resistance may be less closely related to the nature of a specific event or environmental situation than to his ability to deal effectively with that situation. It is often more useful to focus on a patient's attitude, history and coping style than to consider the role of a presumed source of "stress" in the abstract.
5) The fact that levels of steroids, catecholamines, or other measures may change in the course of disease does not necessarily reflect a causal relationship between those measures and host resistance. Although infection is often accompanied by fever, measures taken to reduce the fever do not cure the infection. Similarly, treatment of a frequently ill child with behavioral or pharmacologic "stress-reduction" methods may comprise only symptomatic treatment at best and may indeed cause iatrogenic harm. Enhancing a patient's ability to cope with relevant life events is more likely to be productive in the long run.
6) Psychosocial factors play a role in all phases of disease - onset, treatment, recovery. As Norman Cousins has written, "we need to know more not just about the balancing mechanisms inside the human body, but about the phenomenon of repair and the way stress is related to both."18 Further systematic research on the biological mechanisms underlying psychosocial effects on disease should contribute to a more comprehensive understanding of disease processes, their onset and management.
1. Meyer RJ, Haggcrty RJ; Streptococcal infections in families Factors altering individual susceptibility. Pediatrcs 1962: 29:39.
2. Starfield R, Hankin J. Steinwachs D. et al: Utilization and morbidity: Random or tandem? Pediatrics 1985: 75:241.
3. Plaut SM, Friedman SB: Psychosocial factors in infectious disease, in Ader R (ed): Piychtmeunrnnmunotogy- New York. Academic Press, 1981, pp 3-30.
4. Friedman SB, Glasgow LA: Psychologic rectors and resistance to infectious disease. Pedutr CIm North Am 1966; IM15.
5. Greenfield NS, Roessler R1 Crosley AP: Ego strength and length of recovery from infectious mononucleosis. J Nerv Meni Dis 1959: 128:125.
6. MelamcJ BO, Dearborn M, Hermecz DA: Necessary considerations for surgery preparation: Age And previous experience. Psychusom Med 1983; 45:517.
7. Cunningham AS: Mind. body, and immune response, in Ader R (ed): Psiçhtmeumimrounolugy New York, Academic Press, 1981, pp 609-617.
8. Selve H: The evolution of the stress concept. American Scientist 1975; 61-692.
9. Herd JA: Behavioral factors in the physiological mechanism of cardiovascular disease, in Weiss SM. Herd AJ. Fox BH (eds): ftnpecawrs on Behavioral Mediane. New York. Academic Press, 1981. pp 55-65.
10. Mason JW: Organization of psychoendocrine mechanisms. A review and reconsideration of research, in Greenfield NS, StcmbachRA (eds): Handbook of Psychophystologi. New York, Holt. Rinehart. and Winston. 1972. pp 3-91.
11. Ader R: The role of developmental factors in susceptibility to disease. Im J Psychiatry Med 1974; 5:367.
12. Monjan AA, Collector Ml: Stress-induced modulation of immune response. Science 1977; 196:307.
13. Palmblad J: Stress and immunologic competence: Studies in man, in Ader R (ed): PiycAoneuniimrnunoliigy. New York, Academic Press, 1981, pp 229-257.
14. Kiecolt-Glaser JK, Gamer W. Speicher C, et al: Psychosocial modifiers of immunocompetencc in medical students. Psychatom Med 1984; 46:7.
15. Rogers MP, DubeyD, Reich P: The influence of the psyche and the brain on immunity and disease susceptibility: A critical review. Psychosom Med 1979; 41:147.
16. Rosch PJ: Stress and illness. JAMA 1979; 242:427.
17. Engel GL: The need for a new medical model: A challenge for biomcjicine. Science 1977; 196:129.
18. Cousing N: Stress. JAMA 1979; 242:459.