By Kreipe, Richard E
Somatoform disorders constitute a group of disorders within the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV] that have two features in common: (1) physical (somatic) symptoms that suggest a medical condition, but that are not explained fully by a medical or mental disorder or the direct effects of a substance, and (2) the symptoms “must cause clinically significant distress or impairment in social, occupational or other areas of functioning” [1]. Because of the physical nature of symptoms, somatoform disorders are more likely to present to a primary care provider or medical specialist than to a mental health provider. Six distinct diagnoses are included in this category: somatization disorder, undifferentiated somatoform disorder, conversion disorder, pain disorder, hypochondriasis, and body dysmorphic disorder. Factitious disorder and malingering can present with similar symptoms, and often are considered in the differential diagnosis, but each is placed in a separate category in DSM-IV.
In clinical practice, psychiatric symptoms that are associated with somatoform disorders often are assumed by adolescent patients and their parents to be the result of a physical illness, whereas physicians tend to interpret underlying mental health problems as causing the symptoms. Somatic symptoms present challenges to medical providers who attempt to distinguish a physical source from a psychiatric source, not only because the two domains are inseparably interactive, but also because a diagnostic work-up that is guided by this often proceeds along a path that leads to conflict and dissatisfaction for all parties involved (adolescent patients, their parents, and their physician).
Adolescents present special challenges in the diagnosis and treatment of somatic symptoms that are associated with psychiatric dysfunction. Usually, the second decade of life is characterized by significant changes in biologic, psychologic, and social domains, which blurs the distinctions between function and dysfunction. Primary and specialty care providers for adolescents recognize the significant interactions across these domains in all adolescents, but may not be equipped to meet the challenges that patients present when psychologic problems rise to the level of a psychiatric diagnosis as they do in somatoform disorders. As noted by Spratt and DeMaso [2], the somatoform disorders represent the extreme end of a continuum of somatic symptoms (the other end of which are the mild and self-limited symptoms of unknown etiology that frequently present to primary care physicians) with recurrent complaints that present puzzling diagnostic dilemmas and even more difficult treatment barriers. Adding to the difficulty, after being informed that no medical/surgical illness has been found and that the symptoms cannot be explained fully by a medical diagnosis, adolescents or their parents may continue to seek repeated medical/ surgical evaluation and treatment.
This article addresses the shortcomings of commonly used diagnostic categories and therapeutic strategies, counterbalanced by the advantages of the biopsychosocial approach in the context of somatoform disorders. It also includes practical suggestions and clinical pearls for primary care providers who evaluate and treat adolescents who have somatic symptoms. Following a description of the present DSM-IV diagnostic algorithm in the somatoform spectrum, a discussion of an alternative approach that applies the biopsychosocial approach to adolescents is described. By offering strategies that can be applied to all adolescents, regardless of their symptoms, this article presents an approach that is especially useful in the diagnosis and management of adolescents who have psychiatric problems, with or without somatic symptoms. Each of the following scenarios represents an actual case that was referred to the author for an adolescent medicine consultation at some point in the course of a somatoform illness. They highlight some potential pitfalls in the management of these challenging conditions, and are used to emphasize the value of the biopsychosocial approach for all of the conditions that are addressed in this issue.
Case scenarios
A 16-year-old girl had been evaluated by an ophthalmologist, an allergist, and an otolaryngologist, and was admitted to the hospital by a neurologist because of unrelenting headaches that made it impossible for her to attend school regularly. She had a history of abdominal pain, vomiting, and diarrhea that were attributed to “food allergies”; chronic fatigue and rheumatic symptoms that were attributed to “fibromyalgia”; and recurrent chest pain that was attributed to “costochondritis.” The adolescent medicine consultant returns to the patient’s room as she is being transported for a repeat electroencephalogram (EEG), now with nasopharyngeal leads. Her mother turns to him and says “I hope that they find something…”
A 12-year-old boy had “seizures” while playing soccer. Treatment for asthma had been initiated 4 months earlier, but without improvement in his wheezing. EEG with video monitoring is normal and pulmonary function tests reveal flow-volume loops with inspiratory flattening, which is characteristic of vocal cord dysfunction.
A 15-year-old girl was evaluated for chronic midepigastric pain. After extensive inpatient testing, she was transferred to a university hospital, where repeat tests remained normal and endoscopy failed to identify the cause of her pain. Adolescent medicine consultation revealed that her estranged father had remarried recently and taught at her school. He had not visited her during hospitalization, but had shown much attention to his stepdaughter when she was ill with abdominal pain. The patient described herself as “much more mature than my classmates” and as a “drama queen.” Regular outpatient visits with the adolescent medicine specialist, using a somatically oriented cognitive behavioral approach to symptomatic relief and resolution of ongoing family conflicts, resulted in gradual improvement.
Two years later, she presented to her primary care physician with acute lower abdominal pain. Because of her history of somatoform pain disorder, he focused on stressors and learned that she had not resolved her conflict with family members, but was planning to move to New York City to become an actress. Brief palpation of her abdomen revealed tenderness, but she seemed to “overreact” to the physical examination. She was prescribed antacids, and was recommended to work on family relationships. Five days later, she presented to the Emergency Department because of worsening pain. When a pelvic examination was performed, 75 mL of pus from a ruptured pyosalpinx spontaneously drained from her posterior vaginal vault.
A 17-year-old boy was evaluated for recurrent back pain that was worse at night, but which responded quickly to ibuprofen. He was a well-known scholarathlete from a small rural town, who was feeling “overwhelmed” by stress. The best player on a winless football team, he mentioned to his physician that it would be good to have an “excuse” not to be on the team. He said, “Could you give me an excuse, Doc?.. Just kidding.” He reported trouble sleeping because of the pain and “all the things on my mind.” Physical examination and standard radiographs of the spine were unremarkable; fluoxetine was added to the ibuprofen because of symptoms of depression and anxiety. When the pain did not improve, he was referred to an adolescent medicine specialist for counseling. Because of his history, a CT scan was obtained first, which revealed a suspected small vertebral osteoid osteoma. After this was removed, his pain did not return, but he continued to work on coping with stress, and expressed relief that his pain was not “all in his head,” as he had presumed when prescribed fluoxetine.
Potential pitfalls in diagnosing somatoform spectrum disorders in adolescents
Limitations of the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition in somatoform disorders
The classification scheme that is applied to psychiatric conditions, the DSM, now in its fourth edition and published by the American Psychiatric Association, highlights characteristics or unique features of various mental health diagnoses. Mayou and colleagues [3] recently concluded that the “existing category of somatoform disorders may be regarded to have failed” because (1) the terminology is unacceptable to patients, because interpreting symptoms as merely a mental disorder in somatic form raises doubt about the genuineness of their suffering; (2) dichotomizing symptoms into those that reflect disease or are psychogenic in origin is theoretically questionable, incompatible with cultures that do not dichotomize mind-body issues, and countertherapeutic; (3) somatoform disorders do not form a coherent category and overlap with other psychiatric disorders (eg, depression, anxiety) that include somatic symptoms; and (4) the medical condition exclusion criteria are ambiguous-a patient could be assigned axis III (irritable bowel syndrome) and axis I (undifferentiated somatoform disorder or pain disorder) diagnoses for identical somatic symptoms.
In addition, the DSM-IV has limitations when applied to adolescents who have somatoform disorders because many primary care providers are not familiar with the intrica\cies and subtleties of a formal psychiatric diagnostic evaluation and the application of the five DSM “axes” (I = mental disorders; II = personality disorders and mental retardation; III = physical conditions and disorders; IV = psychosocial and environmental factors; V=global assessment functioning). In addition, most adolescents do not meet full criteria threshold, but fall to an “undifferentiated” or “not otherwise specified” level of specificity within a diagnostic category. Finally, diagnostic criteria for some conditions (eg, somatization disorder), are intended for adult populations, although a pattern of somatic symptoms is established clearly by adolescence.
Diagnostic and Statistical Manual for Primary Care, Child and Adolescent Version
To respond to these and other limitations in the DSM-IV, the American Academy of Pediatrics published a companion document in 1996. The Diagnostic and Statistical Manual for Primary Care (DSM- PC), Child and Adolescent Version takes into account the developmental issues that are considered in pdiatrie primary care [4]. With respect to somatizing and other symptoms that are assumed to have some relationship to emotional issues, the major advantage of this classification scheme is the creation of two hierarchical subthreshold categories for patients who have symptoms: those that do not interfere with everyday functioning, but that might benefit from intervention (somatic complaint variation V65.49), and those that cause distress or impairment (somatic complaint problem V40.3). The formal DSM-IV criteria are retained in DSM-PC for the most symptomatic patients. Some health insurers do not reimburse visits that are labeled only with “V-codes,” although the recognition of such may lead to significant cost savings in the avoidance of unnecessary diagnostic tests. Thus, the DSM-IV remains the standard coding system that is used most commonly in practice and in research; however, rigid adherence to any of the criteria in DSM-IV can be counterproductive clinically. As Morrison noted [5], “diagnoses are not decided by criteria; diagnoses are decided by clinicians, who use criteria as guidelines.”
Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition diagnostic algorithm for somatoform disorders: potential problems at every step
Ruling out underlying medical conditions
The first step in the diagnostic algorithm in somatoform disorders that is suggested in DSM-IV [1] is to rule out an underlying medical condition that might explain the symptoms. This usually involves diagnostic testing that might include: (1) analysis of blood, urine, stool or spinal fluid; (2) indirect imaging by ultrasound, radiographs, CT, or MRI; (3) direct imaging of internal organs by endoscopy or laparoscopy; or (4) procedures that are related to the function of the heart (eg, ECG or echocardiograms), lungs (eg, pulmonary function tests), brain (eg, EEG), metabolism (eg, breath hydrogen), or other targets that are suggested by the symptoms. When presented with symptoms that are not accounted for by history or physical examination, physicians may perform a battery of tests, some of which will be uninformative and unnecessary either “to be complete” and to reassure themselves, or to respond to pressures by the patient or parents to “find out what is wrong.” Because diagnostic tests rarely have 100% sensitivity to detect pathology, some amount of residual uncertainty remains with negative results. Because of false positivity, the more tests that are performed, the greater is the likelihood that some “abnormality” will be detected.
Although such a strategy may satisfy an individual physician’s required level of diagnostic certainty that a medical condition has been “ruled out,” a series of negative tests may lead to requests or demands from the adolescent or parents for additional tests. Often, these consumers arrive at an office visit armed with misinformation that has been gleaned from the Internet. Sometimes the requests are for unusual tests, such as to detect “dysbiosis” that is related to “yeast overgrowth,” which is diagnosed by the presence of antibodies of unknown significance. Often, the sources of such information are for-profit entities (a web address ending in “.com”). Thus, in practice, a physician often is faced with the dilemma of diagnostic testing that confuses, rather than clarifies, “what is wrong.” On the one hand, “absence of proof does not constitute proof of absence. ” Conversely, the presence of abnormal findings (eg, gallstones) does not necessarily account for symptoms (abdominal pain), and the presence of a medical diagnosis may be distracting and cause a psychiatric condition to go unrecognized.
Thus, “normal” findings may not be reassuring, because they are interpreted by the adolescent as “my doctor thinks that nothing is wrong” or “the symptoms are all in my head” (eg, the boy who had osteoid osteoma). This is especially true if a physician follows a report of negative test results (“good news, everything is normal”) with questions about underlying stress or emotional problems that might be causing symptoms. When a physician reports negative test results and then inquires about stress, a linear cause-effect relationship reasonably can be inferred. It is understandable why an adolescent might conclude that the physician believes that symptoms are being “faked,” are due to “emotional instability,” or are used to get “attention.”
Thus, when adolescent patients or their parents seem to be disappointed when they learn of “normal” test results, the reason may be iatrogenic. When diagnostic tests are preceded by a statement such as, “we will do some tests to see what is wrong,” the implication is that the tests will result in a diagnosis. From the patient’s perspective, normal tests mean that the diagnosis remains unknown and the prognosis is uncertain, so the logical next step would be to obtain more tests. Negative test results also are distressing because adolescents may infer that without a positive test, there will not be any effective treatment and the symptoms may continue indefinitely. This only exacerbates the suffering that is being caused by the symptoms. Such was the case with the 16-year- old girl in the case scenarios, who already had established a pattern of bodily symptoms that was well on its way to somatization disorder.
If the physician indicates that no further studies are indicated because “everything has been normal so far,” the patient or parents often become resentful because the severity of the symptoms is often not perceived as being appreciated by the physician. Likewise, there may be a sense of abandonment when the patient hears statements like “there is nothing more I can do for you,””you will just have to learn to live with it” or, “you need to see a psychiatrist, because I cannot find anything wrong with you physically.” These are additional ways in which a rift in the patient-physician relationship occurs. Obtaining more tests may mend this rift temporarily, but when more negative test results are returned, a vicious cycle has already been set in motion. Some practitioners believe that performing “too many” tests only may reinforce the belief that a medical condition exists, which “feeds into the illness.” Not performing “enough tests” may indicate to the adolescent or parents that the symptoms are not being taken seriously, which often leads to “doctor shopping.” The medical care of patients who have somatoform disorders requires a balance of art and science that can help to avoid some of these pitfalls.
Involuntary versus voluntary (intentional) symptoms
The second step in the DSM-IV diagnostic algorithm is to determine if the adolescent is producing symptoms intentionally. The purpose of this step is to distinguish somatoform disorders from factitious disorder (Munchausen syndrome) or malingering, in which symptoms are produced intentionally to assume the sick role, or for some external gain, respectively. Although the somatoform disorders are characterized by symptoms that are not under voluntary control, attempting to determine if symptoms are produced intentionally tends to be counterproductive in clinical practice; it undermines the trust that is required in a therapeutic partnership between a physician and an adolescent patient. Furthermore, it is nearly impossible to determine if symptoms are being feigned; attempts to do so are inferred as the physician believing that symptoms are being produced to avoid something negative (primary gain), to obtain something positive (secondary gain), for attention, or because of psychologic illness. . i
If an adolescent perceives that his/her physician believes that he/she might be “faking it,” the formation of a therapeutic alliance is extremely difficult. In practice, there is little clinical advantage to identifying symptoms as being under voluntary control in the early phases of assessment. Either side of the clinical algorithm presents a slippery slope. Adolescents who are not intentionally producing symptoms are offended and suffer further because they are not believed, or worse, are believed to be “crazy”; adolescents who are producing symptoms intentionally may become indignant, or escalate or modify their symptoms. This dynamic tends to produce resentment on the part of physicians (suspicious of being duped) and defensiveness on the part of patients (for not being believed).
Additionally, it can lead to splitting of parents, if one parent/ stepparent believes that the adolescent is “making up” symptoms, while the other parent/stepparent believes that a medical condition is “being missed” by the physician. Such was the case with the 12- year-old boy in the case scenarios. Because conversion disorder was suspected before the video-EEG was performed, adolescent medicine consultation was obtained early in the course of the illness, and the normal EEGwas interpreted as reassuring. In addition, because a positive diagnosis of conversion disorder was made, the diagnosis of asthma also was reconsidered and was changed to vocal cord dysfunction, based on inspiratory flow-volume loops. Because of this process, his mother and father were unified in the approach to treatment. Of course, if the diagnosis of factitious disorder is being entertained, it is important to corroborate data from numerous sources, but not to confront the adolescent or parent directly until confirmatory data have been obtained. The production of signs or symptoms to assume the sick role-whether by the adolescent or by a proxy-can be associated with significant harm, including death, and always represents significant psychiatric illness. It must be approached cautiously, and should not be avoided.
Symptoms caused by unresolved conflict or stress
The third step in the DSM-IV diagnostic algorithm is to determine if unconscious, unresolved conflict or stress is related to the symptom pattern. In conversion disorder, stress or unresolved conflict that cannot be dealt with on a conscious level is interpreted to submerge below the level of awareness, but emerge as (is converted into) somatic-often neurologic-symptoms. Thus, the symptom often has symbolic meaning, and reflects a “model” [6]. Classically, primary and secondary gains can be identified that represent something that the adolescent is avoiding and is approaching, respectively. The avoidance-approach dynamic is not unique to conversion dynamics, but it provides homeostatic balance by allowing the adolescent to avoid the source of stress or unresolved conflict (primary) and to approach something desirable, such as attention (secondary). As is true of the other elements in the DSM-IV algorithm, this criterion is difficult to apply clinically because it is not clear how one can determine if psychologic factors are related etiologically to the somatic symptoms.
Similar to the dilemma regarding attempts to determine if symptoms are produced intentionally, there is little advantage to make a definitive cause-effect linkage between emotional distress and physical symptoms in conversion or pain disorder. Recognizing that the proposed dynamic reflects the emergence of somatic symptoms because of stress or conflict that could not be dealt with on a conscious level, it is clinically useless and potentially countertherapeutic to bring these issues to conscious awareness during symptom assessment. Treatment is directed at helping the adolescent to resolve the conflict or develop alternative coping skills and to gain insight into how future problems might cause their bodies to function improperly. A physician who merely points out the link between somatic symptoms and emotional distress does not make affected adolescents competent to deal with underlying problems.
Likewise, the presence of significant stress or unresolved conflict does not mean necessarily that the symptoms are causally related. Although the 17-year-old in the case scenarios had acknowledged not coping well with life stressors that deserved attention, their presence caused his physician to miss the symptoms of osteoid osteoma. The history of somatoform pain disorder in the 15-year-old girl sidetracked her physician, and precluded an interval history and physical examination, which caused her pelvic inflammatory disease to be overlooked.
Failure of the biomedical model in somatoform disorders
Faced with all of these problems, it is understandable that Silber and Pao [7] noted that many clinicians are “baffled by the onslaught of symptoms, become annoyed by the time consumed in caring for patients who are ‘not really sick,’ or feel frustrated by the never-ending recurrent complaints.” Fabrega [8] emphasized that the biomedical paradigm espoused by Western medicine, is in part responsible for this situation, because it reduces a patient’s subjective distress and impairment to objective disease and pathophysiologic processes understood only in terms of biophysical or chemical processes.
The prevalence of these conditions adds to the challenge to clinicians. Somatic complaints are common in primary care [9-13]; recurrent abdominal pain accounts for up 5% of pediatric office visits, and headaches, nausea, chest pain, or fatigue are reported by 10% of adolescents [7]. Some somatizing adolescents have coexisting psychopathology, family conflict, school absenteeism, and increased use of health and mental health services [14], as well as emotional problems [15,16] and persistent physical symptoms later in adulthood [17], which compounds this situation. Therefore, a different paradigm is needed for diagnosis and treatment. The biopsychosocial approach offers many advantages with respect to somatizing adolescents and their families.
The biopsychosocial approach in adolescent medicine
Proposed by Engel [18] in a landmark article in the journal Science in 1977 [18], the biopsychosocial approach offers a means of incorporating existing, as well as emerging and new areas (eg, psychoneuroimmunology), that better explain the cause and treatment of all symptoms that might present in health care settings than does the traditional dualistic approach in which illness is dichotomized into two domains: the body or the mind (Fig. 1). Thus, it “supplements and enriches the discoveries of biomedicine, rather than undermining them” [19]. Central to the biopsychosocial model is an appreciation of the continuum of hierarchical natural systems that always are interacting at any point in time in a patient’s experience of symptoms. Engel noted that “each level of hierarchy represents an organized dynamic whole. . .(with) qualities and relationships distinctive for that level of organization. . .In no way can the methods and rules appropriate for the study and understanding of the cell as cell be applied to the study of the person as person or the family as family” [19]. Furthermore, “each system as a whole has its own unique characteristics and dynamics; as a part it is a component of a higher-level system” [20]; however, “neither the cell nor the person can be fully characterized as a dynamic system without characterizing the larger system(s) (environment) of which it is a part” [20]. Thus, “with the systems hierarchy as a guide, the physician from the outset considers all information in terms of systems levels and the possible relevance and usefulness of data from each level for the patient’s further study and care” [20]. In practice, Frankel and colleagues [21] noted that “a comprehensive understanding of every aspect of care from diagnosis to treatment depends upon an appreciation of both linear and nonlinear processes associated with disease and illness.”
Fig. 1. Biopsychosocial model. (From Engel GL. The clinical application of the biopsychosocial model. Am J Psychiatry 1980;137:537; with permission.)
Another important contribution to the literature was Medical Choices, Medical Chances, a book that was published in 1981 by Bursztajn and colleagues [22], which was based on the fact that the practice of medicine inevitably involves probabilities and uncertainties. The way in which uncertainty is handled by a physician can increase suffering or promote healing. Although often considered distinct and separate, subjective domains (eg, patient- reported history, symptoms) and objective domains (eg, physical examination, laboratory findings) are continuous and interactive in this viewpoint. As Cassell [23] pointed out, regardless of the diagnosis or the underlying emotional issues “the patient does not have the option, nor the interest, to relate things objectively where illness is concerned.” Although the diagnostic process traditionally begins with the history, he also noted that “taking a history is unavoidably and actually an exchange of information” [23]. Therefore, issues of cross-systems relationships, trust, and therapeutic partnerships become central to the suffering that the adolescent and family is experiencing and to the healing process. Nowhere is this truer than with adolescent “psychiatric” issues, not only because of the intrinsically subjective nature of the suffering, but because psychiatric diagnoses tend to be highly stigmatized in many cultures. Within psychiatric diagnoses there probably is no greater risk of doing unintentional harm-as well as the potential for doing good-than in the somatoform disorders. The emphasis of the remainder of this article is on the latter.
The recent emergence of biologic psychiatry, with its focus on neurotransmitters and neural circuitry, also runs the risk of being reductionistic. For example, simplifying adolescent depression to the level of a “chemical imbalance in the brain,” addresses only the molecular contribution to the illness. Although such a framework may minimize stigmatization and enhance the acceptance by patients or parents to use medications that “restore balance,” it ignores the other contexts and systems in an adolescent’s life. This is especially true if a 15-minute medication check with a psychiatrist only targets symptom reduction and medication side effects. As pointed out by Sharf and Williams in the article on psychopharmacology elsewhere in this issue, medication can be a useful tool in the treatment of adolescents who have psychiatric symptoms; however, its optimal effect is realized when it is used as a component of a multi-pronged therapeutic approach. Addressing the molecular systems level is important, but the larger view biopsychosocial approach must be integrated into the other systems that are operating in an adolescent’s life.
Communication in adolescent somatoform disorders
Language in communication
The word “communication” is derived from the Latin communicare (to share, or to have in common). Most adolescents see themselves as having little in common with their physicians, so communicat\ion may be limited. Patients who have somatoform symptoms may have additional difficulties communicating, based on the diagnostic approaches that sometimes are used. If a patient and his/her physician share the goal of understanding the likely cause of symptoms and developing a treatment plan that is likely to restore function and promote healing, then successful outcomes are possible. The best way to establish shared goals is through spoken language.
Spoken language in clinical interactions has two dimensions that are relevant to the patient and the provider: receptive and expressive. Although adolescent patients rarely are aware of these issues, clinicians who treat adolescents must attend to receptive and expressive language for the patient and the provider. To emphasize the importance of this concept, Cassell [23] noted that “spoken language is our most important diagnostic and therapeutic tool, and we must be as precise in its use as is a surgeon with a scalpel.”
During diagnosis, language can be an impediment in working with adolescents who have somatoform symptoms. Several examples were illustrated, but others can be considered. For example, seizurelike activity may be labeled as “psychogenic,” whereas abdominal pain is called “psychosomatic.” Regardless of the root term that follows, many adolescents and their parents regard the prefix “psych” as a negative and nonvalidating term. “I cannot find anything wrong with you” is not helpful on several levels. Depending on where the emphasis is placed (or perceived) in this comment, an adolescent might infer that: (1) another physician might be able to find something wrong, (2) there is nothing wrong, (3) the symptoms are being imagined, (4) the symptoms are being falsified intentionally, or (5) there is nothing that can be done. Silber and Pao [7] noted that the concept of “functional versus organic” misses the mark, because it does not reflect the current thinking that symptoms are the result of a complex interplay of several factors.
Identifying pain as “real”
Traditional wisdom teaches the importance of clinicians acknowledging explicitly to the patient who has somatoform pain disorder their belief that the pain is “real”; however, patients often realize that this is a forced and unnatural response. Adolescents who have a broken arm, appendicitis, or strep throat are not informed that their pain is real; explicitly validating symptoms can have the unintended effect of invalidating them. Patients who are referred to adolescent medicine specialists by organ-based specialists note, not uncommonly, “The doctor who sent me here said that he knows that my pain is real, but I think that he was just saying that and did not really believe it.”
Therein lies a subtle, but critical, point with respect to communication, which is applicable to somatoform disorders. In addition to the actual words of spoken language that are expressed and received in communicating, there is an unspoken need for the words to be authentic and truly express what the physician and the adolescent patient truly believe. Pellegrino [24] pointed out that, “a person becomes a patient when, in his perception of his own existence, he passes some point of tolerance for a symptom or debility and seeks out another person who has professed to help. The patient bears and suffers something, and his expectation is that every act of the physician will be to relieve him of that burden and restore his lost wholeness-which is, incidentally, the meaning of the AngloSaxon word heal.”
“Believing” reported symptoms
Physicians may not believe that the symptoms (eg, pain) that are experienced by a patient who has somatoform disorder are real, when “all of the tests are negative.” A significant amount of time may be invested by clinicians to determine if an adolescent is trying to: (1) avoid something (primary gain), (2) seek attention (secondary gain), or (3) feign symptoms for an internal or external reward. Although adolescents who have somatoform disorders may lack interceptive awareness, they usually are extremely sensitive and aware of any lack of authenticity. Scrupulous attention to honesty requires the physician to acknowledge uncertainty.
Therefore, beliefs about the symptoms are more important than what is spoken about the symptoms; acknowledging that pain is real can cause an adolescent to suffer more if the words expressed say “1 believe you,” but the message received says “I do not believe you.” As a basic principle, Cassell [23] suggested that physicians “accept that physical symptoms always have a physical basis no matter what the underlying cause.” Understanding the significance and meaning of symptoms is important in the diagnosis and in planning treatment.
Changing our understanding of symptoms based on scientific knowledge
Dysmenorrhea provides an excellent example of the limitations of the language and our understanding of adolescent somatic symptoms and their relationship to psychologic issues. Between 1972 and 1989, a highly respected quarterly pediatric series published three articles on adolescent dysmenorrhea. In 1972, primary dysmenorrhea was defined by Sloan [25] as “the presence of painful menstruation in the absence of any somatic or pelvic lesion. The term intrinsic. . .was derived from the belief, sometimes still held, that the etiology for the pain lay in the uterus itself.” Based on this understanding, he stated “the cause. . .that can account for the vast number of cases is and remains psychogenic. As stated, this is almost 100% true in the patient in her teens or younger” [25].
By 1981, advances in our understanding of menstrual physiology led Gantt and McDonough [26] to define primary dysmenorrhea as “painful menstruation without significant pelvic pathology.” Citing scientific findings over the previous 5 years based on microtransducer techniques, and studies of pain receptor physiology and a variety of prostaglandins and their metabolites, they concluded that “the common denominator in most women with primary dysmenorrhea is excess myometrial activity” and “the importance of psychological factors in dysmenorrhea is dubious” [26]. In the 1972 article, the focus was on the patient almost to the exclusion of the pain, whereas in the 1981 article, the focus was more on the pain, with less attention on the patient.
A more balanced approach than either of these dualistic conceptualizations was offered in 1989 by Coupey and Ahlstrom [27], who synthesized knowledge about the physiology and psychology of dysmenorrhea and noted that “some adolescent girls will be found to have disability related to menstrual cramps that seems out of proportion to the severity of the pain. Occasionally, an underlying psychosocial problem,. . .personal or family problem may be contributing to the decreased pain tolerance and heightened anxiety centered around the menses in these girls.” This formulation considers the pain in the context of the patient and the various domains that affect pain.
In addition to spoken language, the astute diagnostician interprets the body language of the adolescent patient as part of the evaluation, and makes use of body language and other techniques to facilitate the adolescent’s expressive language. These techniques include: (1) giving the patient undivided attention (writing as little as possible and making direct eye contact), (2) using reflective language to clarify traditional elements of the history (eg, timing, location, radiation, quality, severity, precipitants, and relievers of pain as well as any associated symptoms), (3) using a conversational style to identify things that the patient is unable to do as a result of the symptoms, (4) developing a differential diagnosis that include specific diagnoses or diseases that the patient (and parents) might be worried about. All of these are directed at helping the clinician to understand the story and the meaning of the illness, as well as the suffering that is caused by it. As Cassell [23] pointed about, “thinking about symptoms, attaching meaning to them, searching for explanations, are a much a part of the illness as are its physical expressions. These thoughts are not caused by the illness, they are part of the illness.”
Using reflective listening to establish a therapeutic partnership
Reflective listening goes beyond not talking. It requires being attentive to subtle nuances of verbal and body language. The physician should not expect to get “facts” from the patient and parents, but an “understanding of what it is like for the patient to be sick.” Central to this process should be an exploration of the fears, worries, or concerns that might be evoked by the symptoms- for the adolescent or the parents. Because what may be stressful to one adolescent may be inconsequential to another, it is important not to make assumptions. There are two major fears that adolescents who have somatoform symptoms often have: an unrecognized physical illness or unrelenting symptoms. The somatizing 16-year-old in the case scenarios had both. These can lead to adolescents seeing themselves as vulnerable, and they may be “overprotected” by parents. As Epstein and colleagues [28] noted, in determining any previous assessment or treatment, the emphasis needs to be on the “experience of the illness.” One also must be aware that circumstances change over time. The 15-year-old girl who had somatoform pain disorder had initiated unprotected intercourse and had developed pelvic inflammatory disease.
Nonverbal communication
Communicating with adolescents who have somatic symptoms need not occur only through spoken or body language. Specifically, asking the patient to keep a detailed daily journal that includes all activities and symptoms can provide written documentation of patterns that are impossible to capture in conversation. This also emphasizes to the adolescent that the symptoms are being taken seriou\sly. Alternatively, patients may find it easier to communicate about their illness through poems or drawings. All of these nonverbal formats serve the dual purpose of aiding in the diagnostic process by adding dimensions that usually are not included in traditional medicine, and being therapeutic, because they give the patients an opportunity to express themselves by a means other than through somatic symptoms.
Applied principles of the biopsychosocial approach in adolescent somatoform disorders
There is not a large evidence base regarding the treatment of somatoform disorders in adolescence. In the literature, the sequence of events often is described as assessment, followed by diagnosis, followed by treatment. In practice, these are not discrete processes, but are highly iterative. Rather than address the specific diagnoses separately as occurs in the DSM-IV, the approach that is used by family physicians [29,30] is more applicable in the primary care of adolescents. Campo and colleagues [9,14,15,31] have translated what is known about adult somatoform disorders into pediatrics; their contributions form the basis of many of the principles regarding the management of somatizing adolescents that follow.
Fig. 2. Somatoform disorders: functional gastrointestinal disorders as a model. (Modified from Drossman DA, editor. Rome II: The functional gastrointestinal disorders. McLean (VA): Degnon Associates; 2000. p. 4; with permission.)
The functional gastrointestinal disorders as a model
The gastrointestinal (GI) system is among the most vulnerable to dysfunction, based on psychosocial influences (Fig. 2). Thus, the criteria for the diagnosis of somatization disorder includes two GI symptoms other than pain. Recognizing the profound interaction between the gut and the brain, the biopsychosocial conceptualization was defined clearly and broadly by an international group of GI specialists that developed a compendium, now it its second edition, entitled Rome II: The functional gastrointestinal disorders: diagnosis, pathophysiology and treatment: a multinational consensus [32]. This 764-page monograph focuses on helping the clinician and investigator to: (1) make a positive diagnosis of disorders that result in pain, nausea, vomiting, diarrhea, constipation, or any combination of these-after metabolic, infectious, neoplastic, and other structural abnormalities have been excluded; (2) understand the pathophysiology of these conditions; and (3) make effective treatment decisions. Rome II is a model for all other organ-based specialties to address functional symptoms that are central to somatoform disorders.
The gut is one of the most vulnerable organ systems in the body with respect to psychosocial problems causing symptoms. Although several different systems interact with each other, the brain-gut interactions have been studied the most widely. The central and enteric nervous systems seem to function in interdependent processes that are only beginning to be understood. The resultant GI pathophysiology is not detectable with traditional tests (blood, urine, or stool), or imaging techniques, and the perception of sensations is totally subjective. Therefore, the diagnosis of a functional GI disorder relies heavily on history and physical examination, rather than on excluding a medical or surgical illness. The resulting behaviors and the way in which these conditions are managed by clinicians can have an influence-either positive or negative-on an adolescent’s outcomes and quality of life. The similarities between this model, applied to functional GI symptoms, and our present understanding of dysmenorrhea, applied to functional gynecologic symptoms, are striking.
Targeted history to demonstrate attention to detail and identify medical/surgical conditions
Agreeing on the diagnostic process
A first step in the process of exchanging information (mistakenly labeled as “taking” a history) in somatoform disorders is to have all parties (clinician, adolescent, and parent) agree that physical symptoms have physical causes, but that these are due to pathologic or functional processes. Pathologic processes (eg, inflammation, infection, cancer) cause cellular disruption, tissue damage, or anatomic distortions that can be detected by a variety of tests, imaging procedures, or direct visualization with endoscopy. Other pathologic conditions (eg, epilepsy, cardiac conduction defects, lactose intolerance) are detected through procedures, such as EEG, ECG, or breath hydrogen measurement. Pathologic processes require medication, surgery or both, but also may benefit from supportive treatments that are prescribed with functional disorders. Thus, a daily routine, including a healthy diet and good sleep hygiene, physical activity, social supports, and skills to deal with stress, can be helpful. Functional processes can result in symptoms that are just as debilitating as pathologic ones, but are related to disruption of normal physiologic function to a degree that might not be detected by “testing.” Thus, the history and physical examination become essential elements in diagnosis. The example of a muscle spasm in the calf (“charley horse”) can help patients and parents understand what is meant by a functional disorder. This example is useful in the case of parietotemporal headaches or colicky abdominal pain, in which muscle contraction is the cause of the pain, although the resulting degree of distress and social dysfunction may be extreme. Mothers are especially likely to relate to this, because calf muscle spasms are common during pregnancy. The clinician can point out that muscles function by contracting and relaxing. If a muscle contracts and stays that way (spasm), the result is painful and tender to the touch, but any blood test, imaging, or even a biopsy of the muscle would be “normal.” Generally, functional disorders are diagnosed by history and physical examination; testing is performed to exclude conditions that might represent a threat to health or that may require different treatments.
Detailed history focused on the presenting symptoms
When somatoform disorders are being considered, some practitioners focus on psychosocial issues: wanting to avoid “reinforcing that something is wrong,” but to “go where the money is.” Experience suggests that it is better to focus meticulously on “chief complaints.” This does not reinforce illness, but gives a message that the symptoms are being taken seriously and prepares the patient for talking about symptoms, rather than testing for symptoms. The timing, location, radiation, quality, severity, prcipitants, and relievers of pain, as well as any associated symptoms, should be explored. With respect to functional disorders (eg, those that suggest a neurologic disorder, even if the symptoms are not in keeping with recognized neuroanatomy or physiology), the story of the symptoms should be examined in detail.
Although the history should focus on the details of symptoms, medical and psychosocial issues should be examined together; the assessment lays the groundwork for treatment. Thus, when an adolescent reports that her body “hurts all over,” inquiry about “What does that hurt stop you from doing?” puts the symptoms in the context of psychosocial issues. Even if a patient has a serious medical or surgical condition, it is important to know something about how an adolescent patient might cope with such a condition. Hippocrates is credited with the adage “I would rather know what kind of person has a disease, than what kind of disease a person has.” This is less likely with a “vertical” assessment process, in which medical conditions are addressed first, followed by an exploration of psychosocial factors only after there is a lack of medical evidence to account for symptoms. With a “horizontal” assessment, medical symptoms and psychosocial factors are addressed side by side; the latter are considered as a context in which to understand medical diagnoses or as a possible primary cause for symptoms. This approach also will reveal strengths and vulnerabilities that the adolescent and family have, and increase the likelihood of making a positive somatoform diagnosis, rather than one by exclusion.
History between visits essential to establish patterns
A daily journal of symptoms that is kept by the adolescent can help to establish patterns of what makes symptoms worse or better. An emphasis on the latter is useful in the treatment phase, because somatizing patients often report that “nothing makes it better.” If a parent completes a symptom journal for an adolescent, this is evidence that the suffering is extending to family members. If one asks an adolescent to keep such as journal, it is important to review the log during each visit. Otherwise, the patient may assume that there is not a real interest in the symptoms. Significant progress is being made when the adolescent is able to say, “I do not want to talk about what is in the journal anymore, I want to talk about the trouble that I am having with my parents.”
Physical examination focused on symptoms
Regardless of how unlikely it may seem that any abnormalities will be detected based on the history, a physical examination to determine the cause of functional symptoms should occur on the initial visit, as well as on subsequent visits. Patients who have neurologic symptoms should have a neurologic exam, using as many techniques that involve the “laying on of hands” as possible, including indirect ophthalmoscopy, deep tendon reflex testing, cranial nerve evaluation, and evaluation of cerebellar function. On follow-up visits, this can be done quickly; the interval history can be obtained and feedback regarding findings can be given during the examination. Physical examination for adolescents who have suspected somatoform disorders (1) reinforces the diagnosis of a somatoform condition for the physician; (2) identifi\es any changes that might occur over time, acknowledging that a positive diagnosis of a somatoform disorder is never 100% certain; and (3) demonstrates to the adolescent and parents that the illness will be monitored closely, even though this may not include diagnostic testing.
Laboratory and imaging studies
Laboratory and imaging studies should be selected based on the history and physical examination. Somatizing adolescents or their parents may ask for specific tests to be performed (eg, MRI for a headache, upper GI for abdominal pain). Such requests need to be interpreted with care; they may be generated by (mis)information that was obtained from television, the Internet, friends, or family members. Rather than outright rejection of a test as “inappropriate,” one can respond, “We could get that test, but it only would tell us about anatomy. Based on the history, I suspect that the cause of your pain is due to muscle spasms that will not show up on any test. ” If the symptoms and physical examination cause diagnostic uncertainty, the least invasive test that provides necessary information should be obtained. For example, an ultrasound of the abdomen can provide a great deal of detail. When such a test is reported as “normal,” it can be helpful to review the details of the study-in visual (the actual study) and written forms (report)- with the adolescent and parents. Similar to the detailed history and physical examination, this transmits the fact that the diagnostic process is being conducted carefully to identify any pathologic conditions.
Before any diagnostic procedures are performed, it is worthwhile to prepare the adolescent and parents for the expected results, including a positive framing on “negative” studies. One could say, “Based on your history and physical examination, I feel confident that your headaches are due to muscle spasm. The muscles that are most tender are those on the side of your head that attach to your jaw. As I showed you in the anatomy book, they are called the temporalis and masseter muscles. But, you and your parents are still worried that you might have a brain tumor. We can get a CT scan to provide reassurance that there is no tumor. You already know some of things that cause your headaches to get worse, but there may be others that we have not figured out yet. Even before the CT scan comes back normal, we can work on some things that should help reduce your pain and suffering. For one thing, you are not sleeping very well and we know that sleeping problems can cause these muscles to tighten. You also seem to have a lot of worries, which also can cause involuntary contraction of the scalp muscles that attach to the jaw.”
Applied physiology in diagnosis and the development of intervention strategies
As Barsky and Borus [33] pointed out, without an explanatory model for the symptoms, negative findings provide little reassurance to the patient and family. But, as pointed out by Campo and Fritz [9], after a somatoform disorder is the operating diagnosis, the clinician should present the diagnostic impression clearly, frankly, and directly; however, this should be done in a manner that builds a foundation for intervention. Three physiologic explanations can be used to provide an understanding of the likely cause of symptoms, alone or in combination.
Conditioned responses
Conditioned responses are well-recognized patterns that are based on normal physiology, in which a repeated stimulus causes a repeated response. The result is that the response can occur in the absence of the original conditioning stimulus, especially if it is linked with another stimulus. With operant conditioning, symptoms are self- perpetuating. Thus, symptoms that may have been initiated by a virus or other agent, “take on a life of their own.” Most adolescents and parents are familiar with the concepts of conditioned response in the context of training desired animal behaviors. Using this paradigm, symptoms are viewed as undesired conditioned responses, and operant deconditioning can be included as an element of treatment. To avoid the impression that this is something that is being done to the adolescent, one should include the adolescent in the development of the deconditioning strategies that are used for symptom reduction.
Multiple triggers and precipitants
Because mind-body disconnections often lead to polarized viewpoints about symptom causation, patients often interpret any psychologic interpretation as “you think it is all in my head.” To avoid this dichotomization, it may be helpful to discuss with the adolescent how the body has a limited number of ways to respond to various triggers (ie, the muscles of the GI tract can contract or relax; bowel wall spasm or distention each can cause pain); however, many things can trigger a “final common pathway.” Bowel wall spasm causes the same pain, whether it is related to irritable bowel syndrome or to lymphoma. Factors that are mediated through the central nervous system-and therefore, are made worse by anxiety or depression-should be included in the list of triggers that are discussed. This is more realistic and accurate than there being a single “cause” for symptoms, and prepares the adolescent for working on psychologic issues, in addition to incorporating other therapies (eg, diet [including fiber], physical activity, bowel habits, antispasmodic medications, relaxation techniques). One can avoid the dichotomization by asking, “If working on psychological issues can help your symptoms improve by 10%, wouldn’t that be worth it? Now, I believe that your symptoms can improve a lot more than that, but there is no single thing that will provide full relief. It’s a package deal.”
Reflexes
Because adolescents who have somatoform disorders often are sensitive to implications that their symptoms are voluntary, purposeful, or intentional (possibly because of the DSM-IV diagnostic algorithm), it often helps to emphasize that several automatic, involuntary, subcortical “reflexes” are at work and cause symptoms. Function in the autonomie nervous system is based on the balance of sympathetic and parasympathetic tone. Insufficient or excessive activity or imbalance between the two arms can result in symptoms that are related to this dysfunction. The vagus nerve, which carries autonomie efferents from the brain to the heart, lungs, stomach, and intestines, has numerous outputs that can result in symptoms. Most adolescents are familiar with feeling “butterflies” in their stomach when nervous, and all parents are familiar with the gastrocolic reflex of infants.
The explanation for the symptoms that was provided to the 12- year-old who had conversion reaction in the case scenarios was, “The EEG was normal during the movements that looked like a seizure. Therefore, we know that the muscles were not twitching because of any messages coming directly from the brain, but they were twitching when they were not supposed to and you definitely did not want them to. Somewhere along the pathway between the nerves and the muscles, the messages to fire got mixed up. The same thing happens with the muscles that make your vocal cords move. When they are supposed to relax, they remain contracted part-way. That is why you had noisy breathing that sounded like asthma. The pulmonary function test shows the pattern of what happens if someone tries to breathe in when the vocal cords are not relaxed completely. Fortunately, the muscle twitching and the vocal cord dysfunction are not caused by anything permanent, and should get better over time. But, we do not have to just wait until your symptoms get better; there are some things that we can help you with to get those muscles contracting and relaxing the way you want them to, when you want them to. Some of my patients who have symptoms like these worry about them happening around friends or in school. So, we also need to work on helping you deal with worries.”
Prescribe face-saving interventions
The biopsychosocial approach that has been mapped out so far is highly interactive, and is a therapeutic partnership between the clinician and the adolescent. Underlying mood disturbance, anxiety, stress, or unresolved conflict need to be addressed. Because the presentation is somatic, the treatment must include some kind of somatic intervention, usually in combination with a variety of cognitive-behavioral interventions, such as described by McCann and LeRoux elsewhere in this issue. Family therapy also can be helpful [34], especially if the symptoms are disruptive to family functioning. Mental health treatment may be accepted more readily if it is described as a part of the treatment plan that is included in all such cases, and that the purpose of mental health intervention is to prevent secondary problems that might perpetuate symptoms. Framing the therapist as a professional who will help the clinician manage the overall care also can be helpful. It should be made clear that the primary care or specialty provider will continue to monitor the adolescent’s symptoms in partnership with the mental health therapist. Depending on the nature of the symptoms, the somatic interventions that are prescribed may be attention to daily structure of eating and sleep; physical, occupational, or speech therapy; or various complementary and alternative medicine (CAM) therapies (eg, biofeedback, yoga, massage, acupuncture). It can be useful to include the family in this phase of symptom management, because the family (especially grandparents) may have beneficial folk remedies that could be suggested. If physical, occupational, or speech therapy is prescribed, or if any CAM therapies are pursued, it is useful for the clinician to describe the situation about the adolescent to the therapists to avoid team splitting. Return to school is important because not attending school may cause the adolescent to fall behind academically. It may \be useful to have the patient return to school only after a prescribed period of tutoring to ensure that the patient is caught up with class work. Tutoring should not be open-ended, but should be based on the tutor’s assessment of the time that is needed. Some schools allow in- school tutoring, which provides a graduated re-entry into school; the adolescent is in the school building, but does not have the pressure of being in the classroom. The socialization that can occur in such circumstances can encourage the full return to school. Returning to school half-days initially also can smooth the re- entry process for adolescents who have missed a great deal of school because of somatic symptoms. The plan for school return should be previewed with the school nurse and guidance counselor to ensure that it is feasible and that someone will be able to monitor progress. Because of the dynamics of overprotectiveness that can arise in these situations, a parent may impede return to school. When such collusion is suspected, it may help to contact the other parent directly to engage his or her help in carrying out the plan.
Summary
Somatoform disorders are presented in the first article in this issue of Adolescent Medicine Clinics because the physical symptoms that cause the adolescent to present for diagnosis and treatment reflect the interaction of the psyche and the soma in ways that are poorly understood. Because of dualistic conceptualizations that are encouraged by technology such as MRI, CT scans and other technologically advanced tools, patients who have these conditions often suffer. As noted by Cassell [35], “suffering is experienced by persons, not merely by bodies, and has its source in challenges that threaten the intactness of the person as a complex social and psychological entity.” Clinicians who care for adolescents who have somatic complaints also suffer when they are unable to provide relief of an adolescent’s suffering. Cassell [36] noted that “physicians are less skilled at what were once thought to be the basic skills of doctors-discovering the history of an illness though questioning and physical examination, and working toward healing the whole person.” The biopsychosocial approach offers a means of working toward healing the whole person, and the focus of this article is on practical solutions to difficult challenges that are presented by adolescents and their families.
References
[1] American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th edition. Washington, DC: American Psychiatric Association; 2000.
[2] Spratt EG, DeMaso D. Somatoform disorder: somatization. Available at: http://www.emedicine. com/ped/topic3015.htm. Accessed October 15, 2005.
[3] Mayou R, Kirmayer LJ, Simon G, et al. Somatoform disorders: time for a new approach in DSM-V. Am J Psychiatry 2005; 162:847-55.
[4] Wolraich ML, Felice ME, Drotar D. The classification of child and adolescent mental diagnosis in primary care. In: Diagnostic and statistical manual for primary care (DSM-PC) child and adolescent versions. Elk Grove Village (IL): American Academy of Pediatrics; 1996.
[5] Morrison J. DSM-IV made easy: the clinician’s guide to diagnosis. New York: Guilford Press; 1995.
[6] Gold MA, Friedman SB. Conversion reactions in adolescents. Pediatr Ann 1995;24:296-306.
[7] Silber TJ, Pao M. Somatization disorders in children and adolescents. Pediatr Rev 2003;24(8): 255-64.
[8] Fabrega H. The concept of somatization as a cultural and historical product of Western medicine. Psychosom Med 1990;52:653- 72.
[9] Campo J, Fritz S. Somatization in children and adolescents. J Am Acad Child Adolesc Psychiatry 1994;33:1223-35.
[10] Garralda ME. A selective review of child psychiatric syndromes with a somatic presentation. Br J Psychiatry 1992;161:759- 73.
[11] Alfven G. Preliminary findings on increased muscle tension and tenderness, and recurrent abdominal pain in children. A clinical study. Acta Paediatr 1993;82:400-3.
[12] Garber J, Walker LS, Zeman J. Somatization symptoms in a community sample of children and adolescents: further validation of the children’s somatization inventory. Psychol Assess 1991;3:588- 95.
[13] Oster J. Recurrent abdominal pain, headache, and limb pains in children and adolescents. Pediatrics 1972;50:429-36.
[14] Campo JV, McWilliams L, Comer D, et al. Somatization in pediatric primary care: association with psychopathology, functional impairment, and use of services. J Am Acad Child Adolesc Psychiatry 1999;38:1093-101.
[15] Campo JV, DiLorenzo C, Chiappetta L, et al. Adult outcomes of pediatric recurrent abdominal pain: do they “just grow out of it?” Pediatrics 2001;108:el-7.
[16] Hotopf M, Carr S, Mayou R, et al. Why do children have chronic abdominal pain, and what happens to them when they grow up? Population-based cohort study. BMJ 1998;316:1196-200.
[17] Walker LS, Quite JW, Duke M, et al. Recurrent abdominal pain: a potential precursor of irritable bowel syndrome in adolescents and young adults. J Pediatr 1998;132:1010-5.
[18] Engel GL. The need for a new medical model: a challenge for biomedicine. Science 1977; 196: 129-36.
[19] Frankel RM, Quill RE, McDaniel SH. The future of the biopsychosocial approach. In: Frankel RM, Quill TE, McDaniel SH, editors. The biopsychosocial approach: past, present, future. Rochester (NY): The University of Rochester Press; 2003. p. 4-5, 259.
[20] Engel GL. The application of the biopsypsychosocial model. In: Frankel RM, Quill TE, McDaniel SH, editors. The biopsychosocial approach: past, present, future. Rochester (NY): The University of Rochester Press; 2003. p. 6.
[21] Frankel RM, Quill RE, McDaniel SH. Introduction to the biopsychosocial approach. In: Frankel RM, Quill TE, McDaniel SH, editors. The biopsychosocial approach: past, present, future. Rochester (NY):
Comments