Kurt Kroenke, MDab*, Judith G.M. Rosmalen, PhD, MAc'd aDepartment of Medicine, Indiana University School of Medicine Regenstrief Institute, RG-6, 1050 Wishard Boulevard, Indianapolis, IN 46202, USA cDepartment of Psychiatry, University Medical Center Groningen, University of Groningen, Hanzeplein 1, 9713 GZ Groningen, The Netherlands dGraduate School for Experimental Psychopathology, The Netherlands
The epidemiology of physical symptoms has been reviewed recently [1,2]. Symptoms account for more than half of all outpatient encounters or, in the United States alone, nearly 400 million clinic visits annually. About one half of these are pain complaints (eg, headache, chest pain, abdominal pain, joint pains), one quarter are upper respiratory (eg, cough, sore throat, ear or nasal symptoms), and the remainder are neither pain nor upper respiratory symptoms (eg, fatigue, dizziness, palpitations). About three fourths of outpatients who present with physical complaints experience improvement within 2 weeks, whereas 20% to 25% are chronic or recurrent [3,4]. Symptoms that are self-limiting (viral respiratory illnesses) or explained readily (eg, angina pectoris in the patient who has classic symptoms and known cardiovascular risk factors, or asthma in the patient who has acute dyspnea and wheezing) are not particularly perplexing. Rather, it is the symptoms that are unexplained and often chronic that are frustrating for providers and patients and costly for the health care system.
This article highlights (1) an overview of unexplained symptoms and so-matization; (2) limitations of the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) in classifying somatoform disorders; (3) predictors of psychiatric comorbidity in patients with physical symptoms; and (4) measuring and managing symptoms.
* Corresponding author. Regenstrief Institute, RG-6, 1050 Wishard Boulevard, Indianapolis, IN 46202.
E-mail address: [email protected] (K. Kroenke).
Studies in the general population as well as primary care and specialty clinic patients have shown that at least one third to one half of symptoms are medically unexplained . One reason for this absence of medical explanation is the frequent lack of objective findings on physical examination or diagnostic testing. Another reason may be the misleading nature of objective findings in certain patients (eg, the cardiac flow murmur that has no causal connection to the patient's atypical chest pain or shortness of breath). Several symptom categories are more descriptive than etiologic. Mechanical low back pain, nonulcer dyspepsia, myofascial pain syndromes, and costochon-dritis are only a few examples of diagnoses with which symptomatic patients are labeled commonly, but for which diagnostic criteria typically are vague. In one of the few instances where this has been studied rigorously, it was found that three experts who reviewed the same data on the same patients and used explicit diagnostic criteria demonstrated only modest interobserver agreement on classifying the cause of dizziness in specific patients . Likewise, the distinction between migraine and tension headache is not always clear-cut, which leads some experts to argue for a continuum theory of migraine on one end of the spectrum and tension headache on the other, rather than a dichotomous classification . This diagnostic heterogeneity among clinicians who evaluate the same symptoms leads to considerable variability in diagnostic testing and treatment approaches .
Proliferation of diagnostic testing has yielded a burgeoning number of false positive results that may be linked mistakenly to nonspecific symptoms. One example is the attribution of low back pain to disc abnormalities that are seen on MRI, a diagnosis that is complicated by the fact that 40% of asymptomatic controls had some degree of disc abnormality on MRI . In fact, radiographic abnormalities in an important fraction of the general population has led to a new term, "incidentaloma" [9,10]. Other examples include the overdiagnosis of Lyme disease in a patient who has fatigue, musculoskeletal pain, and low-level antibody titers [11,12] or "subclinical hypothyroidism'' in a patient who has vague symptoms and borderline elevations of thyroid-stimulating hormone. Meador  warned of the overinterpretation of laboratory (as well as physical) findings in his classic essay, "The Art and Science of Nondisease.''
Experts do not agree entirely on how to define somatization. There are two competing definitions. The first maintains simply that somatization is the process whereby individuals experience and report physical symptoms that, after appropriate investigation, cannot be explained fully by a known general medical condition. The second definition requires the absence of an explanatory medical condition and the presence of psychologic factors that are causing or contributing to the symptom. Confirming that psychologic factors are the actual cause of physical symptoms, rather than merely a consequence or coexisting condition, may not be easy. Also, somatizing patients may resist efforts to attribute their symptoms to nonphysical causes strenuously, which makes it difficult for the practitioner to explore emotional underpinnings openly or offer psychologic treatments.
Although a chronic history of unexplained symptoms builds the strongest case for somatization, follow-up studies of selected symptoms, such as fatigue, dizziness, chest pain, abdominal pain, palpitations, and back pain, have confirmed the clinician's initial judgment that a symptom is unexplained usually is correct and that the delayed emergence of serious diagnoses that were not suspected initially is rare [14-20].
Somatization is associated with increased health care use, functional impairment, provider dissatisfaction, and psychiatric comorbidity. Somatizing patients disproportionately use health care resources, and experience high rates of clinic and emergency room visits, excessive diagnostic testing, frequent subspecialty referrals, and polypharmacy because of multiple therapeutic trials. Moreover, care of the somatizing patient frequently is disjointed because of the number and diversity of health care providers that are involved. This fragmentation of care makes it especially difficult to control medication prescribing, test ordering, and other medical costs.
Medically unexplained symptoms are a major public health problem. The lack of objective findings leads to diagnostic heterogeneity among clinicians who evaluate the same somatic symptom. In case of a chronic history of unexplained symptoms, the presence of a somatoform disorder might be suspected, in which psychologic factors are causing or contributing to the symptoms.
Diagnosis of somatoform disorders: limitations of the Diagnostic and Statistical Manual of Mental Disorders diagnostic system
The first edition of the DSM was published in 1952. Although the first two editions of the DSM were based mainly on psychodynamic etiologic principles, since the Diagnostic and Statistical Manual of Mental Disorders, Third Edition (DSM-III), which was published in the early 1980s, the editions have been based increasingly on clusters of symptoms that were derived primarily from clinical observation and epidemiologic research. Despite this development, criteria for DSM diagnoses are not established on a truly empiric basis; instead, expert consensus is used. Expert consensus remains an opinion rather than the objective truth. Consequently, the DSM diagnostic system has been criticized extensively.
Somatoform disorders form no exception to this rule. The category of so-matoform disorders first was included in DSM-III as a group of diagnoses that was characterized by the presentation of physical symptoms that suggest a medical condition. In DSM-IV, the somatoform category is composed of the subcategories of somatization disorder (SD), undifferentiated somato-form disorder, hypochondriasis, body dysmorphic disorder, conversion disorder, and pain disorder (diagnostic criteria are summarized in Table 1). As
Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition diagnostic criteria for somatoform disorders
Disorders Diagnostic Criteria
Somatization A. A history of many physical complaints beginning before age 30 years that occur over a period of several years and result in R
disorder treatment being sought or significant impairment in social, occupational, or other important areas of functioning. m
B. Each of the following criteria must have been met, with individual symptoms occurring at any time during the course of the E disturbance: &
(1) Four pain symptoms: a history of pain related to at least four different sites or functions (eg, head, abdomen, back, joints, o extremities, chest, rectum, during menstruation, during sexual intercourse, or during urination) A
(2) Two gastrointestinal symptoms: a history of at least two gastrointestinal symptoms other than pain (eg, nausea, bloating, vomiting E other than during pregnancy, diarrhea, or intolerance of several different foods)
(3) One sexual symptom: a history of at least one sexual or reproductive symptom other than pain (eg, sexual indifference, erectile or ejaculatory dysfunction, irregular menses, excessive menstrual bleeding, vomiting throughout pregnancy)
(4) One pseudoneurologic symptom: a history of at least one symptom or deficit suggesting a neurologic condition not limited to pain (conversion symptoms, such as impaired coordination or balance, paralysis, or localized weakness; difficulty swallowing or lump in throat; aphonia; urinary retention; hallucinations; loss of touch or pain sensation; double vision; blindness; deafness; seizures; dissociative symptoms, such as amnesia; or loss of consciousness other than fainting)
(1) After appropriate investigation, each of the symptoms in criterion B cannot be explained fully by a known general medical condition or the direct effects of a substance (eg, a drug of abuse, a medication)
(2) When there is a related general medical condition, the physical complaints or resulting social or occupational impairment are in excess of what would be expected from the history, physical examination, or laboratory findings
D. The symptoms are not feigned or produced intentionally (as in factitious disorder or malingering).
Undifferentiated A. One or more physical complaints (eg, fatigue, loss of appetite, gastrointestinal or urinary complaints) somatoform B. Either (1) or (2):
disorder (1) After appropriate investigation, the symptoms cannot be explained fully by a known general medical condition or the direct effects of a substance (eg, a drug of abuse, a medication)
(2) When there is a related general medical condition, the physical complaints or resulting social or occupational impairment is in excess of what would be expected from the history, physical examination, or laboratory findings
C. The symptoms cause clinically significant distress or impairment in social, occupational, or other important areas of functioning
D. The duration of the disturbance is at least 6 months
E. The disturbance is not better accounted for by another mental disorder (eg, another somatoform disorder, sexual dysfunction, mood disorder, anxiety disorder, sleep disorder, or psychotic disorder)
F. The symptom is not produced or feigned intentionally (as in factitious disorder or malingering)
Hypochondriasis A. Preoccupation with fears of having, or the idea that one has, a serious disease based on the person's misinterpretation of bodily symptoms
B. The preoccupation persists despite appropriate medical evaluation and reassurance
C. The belief in criterion A is not of delusional intensity (as in delusional disorder, somatic type) and is not restricted to a circumscribed concern about appearance (as in body dysmorphic disorder)
D. The preoccupation causes clinically significant distress or impairment in social, occupational, or other important areas of functioning.
E. The duration of the disturbance is at least 6 months
F. The preoccupation is not accounted for better by generalized anxiety disorder, obsessive-compulsive disorder, panic disorder, a major depressive episode, separation anxiety, or another somatoform disorder
Body dysmorphic A. Preoccupation with an imagined defect in appearance. If a slight physical anomaly is present, the person's concern is markedly disorder excessive.
B. The preoccupation causes clinically significant distress or impairment in social, occupational, or other important areas of functioning
C. The preoccupation is not accounted for better by another mental disorder (eg, dissatisfaction with body shape and size in anorexia nervosa)
(continued on next page)
Table 1 (continued)
One or more symptoms or deficits affecting voluntary motor or sensory function that suggest a neurologic or other general medical condition
Psychologic factors are judged to be associated with the symptom or deficit because the initiation or exacerbation of the symptom or deficit is preceded by conflicts or other stressors
C. The symptom or deficit is not produced or feigned intentionally (as in factitious disorder or malingering)
D. The symptom or deficit cannot, after appropriate investigation, be explained fully by a general medical condition, or by the direct effects of a substance, or as a culturally sanctioned behavior or experience
E. The symptom or deficit causes clinically significant distress or impairment in social, occupational, or other important areas of functioning or warrants medical evaluation
F. The symptom or deficit is not limited to pain or sexual dysfunction, does not occur exclusively during the course of somatization disorder, and is not better accounted for by another mental disorder
A. Pain in one or more anatomic sites is the predominant focus of the clinical presentation and is of sufficient severity to warrant clinical attention
The pain causes clinically significant distress or impairment in social, occupational, or other important areas of functioning Psychologic factors are judged to have an important role in the onset, severity, exacerbation, or maintenance of the pain
The symptom or deficit is not produced or feigned intentionally (as in factitious disorder or malingering) The pain is not accounted for better by a mood, anxiety, or psychotic disorder and does not meet criteria for dyspareunia
Adapted from the American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders DSM-IV, 2004; with permission.
is the case for other DSM diagnostic categories, the overall concept of somatoform disorders and its subcategories have remained a focus of criticism despite subsequent modifications [21,22]. The main problem of DSM seems to be the implicit assumption that mental disorders are disease entities that can be defined by operationalized sets of symptom criteria; however, this assumption is questionable and comes with potentially harmful side effects. Three of these side effects are discussed: problems identifying the border between disorder and normality, proliferation of new diagnoses, and magnification of comorbidity, thereby focusing on the diagnosis of somatoform disorders.
It is unclear how to distinguish between mental disorder and nondisor-der problems of living. The main assumption of a categorical diagnostic model—that diagnoses have an identifiable separation from normality— implies that the distribution of clinical features of the phenotypes of diseases states must have discontinuities; however, many of the DSM diagnoses most likely represent entities in which cut-off points have been superimposed arbitrarily on variables that seem to be distributed continuously in the population. For example, a variety of studies has found that personality disorders merge with normality . Despite the fact that researchers have been unable to identify a qualitative distinction between normal functioning and mental disorder, DSM-IV provides specific and explicit rules for the number and severity of sign and symptom criteria that are required for diagnosing a "case" of mental disorder. The thresholds for diagnosis that are provided in DSM-IV remain largely unexplained and weakly justified .
Somatoform disorders are no exception to this shortcoming of DSM. Escobar and colleagues  reviewed the criteria for the diagnosis of SD in the subsequent DSM editions: from 25 unexplained somatic symptoms from a list of 59 in addition to attitudinal features in DSM-I; to a symptom count of 14 (men, DSM-III) or 16 (women, DSM-III) followed by 13 (both genders, DSM-IIIR) from a list of 37 symptoms; to the current criteria of 8 symptoms coming from four designated organ systems in DSM-IV. The modifications already indicate the arbitrary distinction between a case and a noncase. As observed for other psychiatric phenotypes, the number of somatic symptoms that a person reports is distributed continuously in the general population, and the diagnosis merely represents an extreme on a continuum of distress . There seems to be no qualitative difference between patients who meet the criteria for the diagnosis of SD, and those who only have some symptoms but do not meet the formal criteria. Even the diagnostic criteria of disorders that are not based on symptom counts, are, by definition, arbitrary. Consider, for example, hypochondriasis: no norms exist for levels of illness worry that are appropriate for a patient's medical condition .
The arbitrary criteria for distinguishing between the presence and absence of a mental disorder obviously result in two main diagnostic problems: false positives and false negatives. Because symptoms of many mental disorders commonly occur in persons who do not have a psychiatric disorder as normal reactions to psychosocial stress, avoiding false positives is difficult. The solution for this problem is the inclusion of the clinical significance criterion (eg, "the disturbance causes clinically significant distress or impairment in social, occupational, or other important areas of functioning'') to a large number of disorders in DSM-IV. The clinical significance criterion attempts to minimize false positive diagnoses in situations in which the symptom criteria do not necessarily indicate pathology. This criterion has been criticized on several grounds . The main criticism is that many of the diagnostic symptom criteria are associated inherently with significant impairment, so that the clinical significance criterion is redundant, and therefore, does not affect the definition of a case. In case of somatoform disorders, a study in a primary care setting demonstrated that when the new DSM-IV criterion of moderate to severe clinical impairment was ignored, the prevalence of so-matoform disorders increased from 16% to 22% . This indicated that the clinical significance criterion might avoid false positive diagnoses of so-matoform disorders.
The clinical significance criterion increases the chance of false negative diagnoses (ie, failure to diagnose psychiatric disorders that are present), or subthreshold conditions. There is much evidence to support the notion that subthreshold conditions are important public health problems; however, given the problems with clarifying diagnostic boundaries between normality and (suprathreshold) DSM disorders, it is logical that it is even more difficult to define subthreshold diagnoses. Abridged SD and multisomato-form disorder have been proposed as less restrictive somatoform disorders, and both were demonstrated to be associated with excess functional impairment and health care use [26,30].
One of the most striking features of every new edition of the DSM is the addition of a large number of new psychiatric disorders. Fig. 1 shows the numbers of diagnoses included in the DSM editions: 106 in DSM-I (1952), 182 in DSM-II (1968), 265 in DSM-III (1980), 292 in DSM-IIIR (l987), and 365 in DSM-IV (1994). There is a close relationship between the year of publication of a DSM edition and number of diagnoses that is included in that edition (in which year of publication explains as much as 98% of the variance in number of diagnoses). By extrapolating this relationship to DSM-V, which is expected to be published in 2012, it can be estimated that this edition will include 457 diagnoses.
1 150 E
1955 1965 1975 1985 1995 year of publication
1955 1965 1975 1985 1995 year of publication
Fig. 1. Number of diagnoses of psychiatric disorders in DSM editions as a function of year of publication.
It is questionable whether the new diagnostic entities that are added to every new edition reflect an enhanced understanding of the etiology of psychiatric disorders. In case of somatoform disorders, most of the subcate-gories remain controversial. Several of the somatoform subcategories also could be regarded as specific manifestations of psychiatric diagnoses from other diagnostic categories. For example, based on the phenomenological similarities with several anxiety disorders , it could be argued that hypo-chondriasis should be regarded as an anxiety disorder that happens to focus on health issues . In the same line, body dysmorphic disorder merely might be a form of obsessive-compulsive disorder in which the compulsive ideas and behaviors (mirror checking) and obsessive thoughts concern the body . Conversion disorder could be reunited with dissociative disorder, because similar psychologic processes seem to underlie these disorders, despite their descriptive differences . SD might be better regarded as a combination of personality disorder and anxiety and depressive syndromes . Significant difficulties were found in differentiating between the diagnoses of undifferentiated somatoform disorder and pain disorder in a sample of 127 patients . Independent of the major presenting complaint, most of the patients had multiple pain and nonpain symptoms. The investigators concluded that the differential diagnosis relies on whether the investigator judges the number of symptoms or the preoccupation with pain as the most important factor for diagnosis . In conclusion, it is unclear whether the existing somatoform disorders truly are independent diagnostic entities.
Additionally, despite this proliferation of presumed psychiatric diagnostic entities, DSM diagnoses still do not cover common clinical cases adequately. To solve this problem, DSM includes "not otherwise specific'' (NOS) categories, into which atypical conditions or conditions that do not meet the full criteria for a specific mental disorder are placed. Undifferentiated somatoform disorder was placed alongside somatoform disorder-NOS in DSM-IIIR to diagnose patients who had a somatoform disorder that did not fit into one of the specific DSM-III categories . Subsequent studies demonstrated that undifferentiated somatoform disorder was the most common somatoform disorder , which illustrates the limitations of the existing DSM diagnostic categories.
Given the proliferation of psychiatric diagnoses, it is not surprising that psychiatric comorbidity also is encountered frequently. High rates of psychiatric comorbidity have been observed in numerous clinical and epidemio-logic samples [35-39].
As can be anticipated, somatoform disorders also are accompanied frequently by other mental disorders. A high comorbidity of somatoform disorders and especially anxiety or depressive disorders was a common finding in previous studies [25,40]. A recent study in a primary care setting demonstrated that 26% of patients who had a somatoform disorder also had an anxiety or depressive disorder. Conversely, 54% of all patients who had an anxiety or depressive disorder also had a somatoform disorder . Several investigators have studied psychiatric comorbidity for the specific sub-categories of somatoform disorders in the community, primary care patients, and psychiatric patients. Concerning SD, depressive disorders have been diagnosed in 55% to 94% of patients who have SD, panic disorder has been diagnosed in 26% to 45% of patients who have SD, and phobic disorder has been diagnosed in 25% to 39% of patients who have SD ; the higher rates usually are found in psychiatric patients. Of patients who had SD in the primary care setting, 34% met the diagnostic criteria for generalized anxiety disorder . Additionally, excessive co-occurrence of personality disorders has been found; structured interviews documented that 23% of patients who had SD in the primary care setting had one personality disorder and 37% had two or more personality disorders . The four most frequently identified personality disorders were avoidance (27%), paranoia (21%), self-defeating (19%), and obsessive-compulsive (17%) . Studies in primary care outpatients who had hypochondriasis demonstrated that 62% to 88% had one or more additional clinical psychiatric disorder; the overlap was greatest with depressive (44%-55% for any depressive disorder) and anxiety (22%-86% for any anxiety disorder) disorders . Likewise, patients who have body dysmorphic disorder have a current prevalence of 59% for major depression and a lifetime prevalence of 83%; they also have a 35% lifetime prevalence for primary social phobia and an 11% lifetime prevalence for panic disorder .
This degree of overlap exists despite the fact that several DSM diagnoses include hierarchical criteria to ensure that diagnoses are mutually exclusive. Consider, for example, the DSM-IV definition of generalized anxiety disorder. The essential feature of this disorder is excessive anxiety and worry that occurs for a certain period of time about several events or activities. The diagnostic criteria pose restrictions to the focus of the anxiety and worry; it should not be confined to features of another clinical psychiatric disorder, such as having a panic attack (eg, in panic disorder), being embarrassed in public (eg, social phobia), being contaminated (eg, obsessive-compulsive disorder), being away from home or close relatives (eg, separation anxiety disorder), gaining weight (eg, anorexia nervosa), having multiple physical complaints (eg, SD), or having a serious illness (eg, hypochondria-sis), and the anxiety and worry should not occur exclusively during post-traumatic stress disorder. Without these exclusion criteria, the overlap between somatoform disorders and other DSM disorders would be even more extensive.
In the case of somatoform disorders, there is a special problem concerning the exclusion criteria; it is unclear whether functional somatic syndromes, such as chronic fatigue syndrome or fibromyalgia, count as exclusions. This may result in artificial comorbidity because patients may be classified as having a nonpsychiatric medical disorder (eg, irritable bowel syndrome) and a somatoform disorder (eg, undifferentiated somatoform disorder or pain disorder) for the same somatic symptoms . Several investigators concluded that a substantial overlap exists between the individual functional somatic syndromes, and that the similarities between them (in case definition, reported symptoms, and in nonsymptom association [eg, patients' sex, outlook, and response to treatment]) outweigh the differences [43-45]. It seems that the difference between somatoform disorders and functional somatic syndromes depends mainly on the investigator (functional somatic syndromes are used widely in primary care and general medicine, whereas somatoform disorders are used in psychiatric settings).
A second diagnostic problem is the criterion excluding symptoms that are "better accounted for by a general medical condition.'' Evidence from multiple studies indicate that even symptoms that might be attributable to a co-morbid physical disorder may not be explained fully by that disorder. Examples include
Exercise tolerance in some patients who have chronic obstructive pulmonary disease may correlate better with depression than with abnormalities on spirometric testing .
Diabetic symptoms may correlate more strongly with psychologic factors than with the degree of glucose control [47,48].
Palpitations correlate with stress and tendencies to amplify somatic symptoms, rather than cardiac ectopy, on electrocardiographic monitoring .
Angina burden in patients who have heart disease is predicted more by depression severity than by findings on echocardiographic stress testing .
Cognitive complaints following coronary artery bypass surgery correlate better with measures of depression and anxiety than with neuropsychologic test results .
Among patients who have cancer or are HIV positive, unexplained symptoms are common and are associated strongly with depression [52,53].
This "nonspecificity" of apparently disease-specific symptoms was highlighted in a recent study of nearly 3500 patients who were 60 years and older and attended a primary care clinic . As shown in Fig. 2, chest pain was only moderately more prevalent in patients who had heart disease compared with other diseases; this finding also was noted for other conditions (eg, joint pain and arthritis, dyspnea and pulmonary disease).
In summary, the diagnosis of DSM somatoform disorders in patients who have medically unexplained symptoms is complicated. Problems include identifying the border between disorder and normality. Moreover, diagnostic proliferation and high rates of comorbidity among patients who have DSM-diagnosed mental disorders suggest that symptoms that are associated with psychopathology do not divide easily into mutually exclusive categories. The nonspecificity of current therapies illustrates this; although the number of diagnoses continues to increase, the main classes of drugs are used increasingly across diagnostic categories. For example, selective serotonin reuptake inhibitor medications, although initially regarded as anti-depressants, were reported to be part of the treatment of many anxiety disorders, schizophrenia (negative symptoms), and anorexia nervosa [55,56]. The use of atypical antipsychotics is well established in bipolar disorder, schizophrenia, and psychosis in general, but it also has been successful in the treatment of psychotic and behavioral disorders in dementia of various types, refractory obsessive-compulsive disorder, pervasive
Joint Pain Fatigue Dyspnea Chest Pain Dizziness
Fig. 2. ''Nonspecificity'' of apparently disease-specific symptoms, as highlighted in a prevalence study of physical symptoms in 3498 patients who were 60 years and older and attended a primary care clinic. The prevalence of each symptom is shown in patients who have one of five medical conditions: cardiac disease, pulmonary disease, hypertension, diabetes, and arthritis. Arrows denote the disease for which a particular symptom is expected to be more prevalent.
developmental disorder, stuttering and Tourette's syndrome, refractory depression, and borderline personality disorder . The options for psychotherapy are restricted to cognitive-behavioral therapy (and its derivates) and psychodynamic therapy. Thus, new diagnostic categories may not lead to the development of more specific therapies. It is obvious that a mul-tidisciplinary approach is necessary in patients who have somatoform disorders. The first prerequisite is to identify the somatizing patient, and thus, to recognize the predictors of psychiatric comorbidity in patients who have physical complaints.
Depressive, anxiety, and somatoform disorders are the three most common comorbid psychiatric conditions in patients who present with physical complaints [2,58-63]. At least one third to one half of all physical complaints are unexplained medically; a depressive disorder can be diagnosed 50% to 60% of the time and an anxiety disorder can be diagnosed 40% to 50% of the time, regardless of the symptom. Also, physical symptoms are unexplained medically in at least one third of patients who are referred to specialty clinics . Among patients who were referred to three specialty clinics (gastroenterology, rheumatology, neurology), depression was present in one quarter to one third of patients in each clinic, and depressed patients were only about one-fourth as likely to have a physical diagnosis established as an explanation for the symptoms that triggered their referral.
Although the specific type of symptom is not particularly important in terms of predicting depression or anxiety, the number of symptoms is. As shown in Fig. 3, there is a strong relationship between the number of physical symptoms that is endorsed by patients as currently bothersome and the likelihood of a coexisting depressive or anxiety disorder [59,60]. Thus, the number of symptoms might be an appropriate identifier ("red flag'') for patients who require an integrated analysis of their complaints (see the article by Huyse and colleagues elsewhere in this issue). This relationship is for total symptom count—medically explained and unexplained. Although limiting the count to unexplained symptoms strengthens the relationship with depression and anxiety [26,59,62], the total symptom count itself is a powerful predictor and does not require the clinician to go through the time-consuming and subjective process of adjudicating the somatoform nature of each individual symptom. Besides the simple total number of symptoms, other symptom count thresholds that predict psychiatric comorbidity are three or more unexplained bothersome symptoms with at least a several year history of somati-zation , pain in at least two different regions of the body, and multiple functional somatic syndromes .
One common question is "What is the cause of somatic symptoms in the person who has multiple medical conditions?''. Couldn't the person who has
Number of physical symptoms
Fig. 3. Increased likelihood of depressive or anxiety disorder as the number of physical symptoms endorsed on a 15-symptom checklist increased in two primary care studies that involved 500 and 1000 patients.
Fig. 3. Increased likelihood of depressive or anxiety disorder as the number of physical symptoms endorsed on a 15-symptom checklist increased in two primary care studies that involved 500 and 1000 patients.
heart disease, chronic lung disease, esophageal reflux, and osteoarthritis be bothered by multiple physical symptoms? Although this is true theoretically, the reality is that patients who have multiple medical problems typically endorse only a limited number of symptoms as currently bothersome on a symptom checklist; when offered a menu, they focus on a few predominant complaints. Physical symptom count has a much stronger relationship with depressive and anxiety disorders than with the number of medical disorders. A second question is ''What is the possibility of a single polysymptomatic disease in someone who reports multiple symptoms?''. The fact is that depression and anxiety are exceptionally common, whereas these multisystem diseases are infrequent in primary care and usually have other objective manifestations besides symptoms.
Besides unexplained or multiple symptoms, Box 1 summarizes some of the factors that increase the likelihood of a concurrent psychiatric disorder in the individual who has physical symptoms. Chronicity or recurrence of symptoms is one such factor. Also, the clinician's own reaction to the visit is another predictor [66-68]. One out of six primary care visits is perceived as ''difficult'' by the physician; patients who trigger such difficult encounters are three times more likely to have a depressive or anxiety disorder and are nine times more likely to have a somatoform disorder. Four predictors that constitute the ''S4 model'' (recent stress, high symptom count, high symptom severity, and low self-rated health) are assessed easily. The likelihood of a depressive or anxiety disorder with 0, 1,2, 3, or 4 of these S4 predictors is 8%, 16%, 43%, 69%, and 94%, respectively [2,60]. Finally, other predictors of psychiatric comorbidity in patients with physical complaints include high health care use  and medication history, including polypharmacy; failure to respond to numerous medication trials for the same symptom; and intolerance of multiple medications, which Barsky and colleagues  referred to as the ''nocebo'' effect.
Was this article helpful?