Textbook of Psychiatry/Somatoform Disorders

Introduction
The somatoform disorders have in common the "repeated presentation of physical symptoms, together with persistent requests for medical investigations, in spite of repeated negative findings and reassurances by doctors that the symptoms have no physical basis" (World Health Organization, 1992). Psychiatrists worldwide use either the ICD-10 or DSM-IV systems of classification when diagnosing mental illness. For most conditions there is little difference between the two systems, but for the somatoform disorders the conditions included differ slightly – see Table 1 for details.

Table 1: Comparison of Somatoform Disorders: ICD-10 vs. DSM-IV

For the purposes of this chapter we take an inclusive view of disorders loosely grouped under the somatoform label and cover the following conditions: •	Somatization Disorder •	Hypochondriacal Disorder •	Somatoform Pain Disorder and Chronic Pain •	Conversion (Dissociative Motor) Disorder •	Body Dysmorphic Disorder •	Functional Somatic Syndromes (e.g., chronic fatigue syndrome/myalgic encephalomyelitis, fibromyalgia, chronic pelvic pain, multiple chemical sensitivity).

The Somatoform Disorders are important to recognise because they are relatively common, costly and almost invariably present to doctors other than psychiatrists. In addition, many doctors find patients with these disorders difficult to understand and treat. The feature that all of these illnesses have in common is the patient’s experience of medically unexplained symptoms, which refers to physical (or somatic) symptoms that are disproportionate to identifiable physical disease.

Terminology
The terminology is confusing in this area, as many terms are used interchangeably. For example, although we often use the term medically unexplained symptoms in this chapter you may also encounter terms such as "somatization," "functional symptoms" or "hysterical symptoms" seemingly referring to the same thing. It is possible for one patient to fulfil diagnostic criteria for several somatoform disorders at one time (e.g., somatoform pain disorder and dissociative disorder) which has led to criticism of current diagnostic systems, and it is likely that future versions of ICD/DSM will change how such disorders are defined (Kroenke, Sharpe et al. 2007).

To make matters worse, psychiatrists often use different diagnostic terminology to that used by their medical colleagues; these differences can hamper doctors’ ability to come to a shared understanding of a patient’s problems. Take, for example, a woman who suffers from a wide number and range of symptoms for which no adequate pathological cause has been found. These symptoms have been present for many years, have resulted in marked disability and, despite a long history of consultations with many different doctors, there has been no improvement. The woman’s medically unexplained symptoms include fatigue, dizziness, headache, subjective limb weakness and painful joints. A psychiatrist makes a diagnosis of "somatization disorder," whilst a rheumatologist diagnoses "fibromyalgia" and a neurologist "chronic fatigue syndrome/myalgic encephalomyelitis." The patient herself rejects all of these diagnoses and prefers to think of herself as having "multiple chemical sensitivity." In the field of the somatoform disorders, the labels often say more about the specialty of the person applying them than any underlying pathology. The lesson to learn here is that these diagnostic labels are descriptive, often overlapping and seldom uncontentious.

Clinical Symptoms and Classification
All somatoform disorders are highly co-morbid (i.e., co-exist) with each other and with anxiety and depression. Therefore screening for anxiety and depression, which are treatable, should be undertaken in any patient presenting with a medically unexplained syndrome. In the following section we go through the somatoform disorders in turn and highlight their diagnostic features. The diagnostic descriptions are based on ICD-10 criteria where possible. We emphasise from the outset that the classification of the so called somatoform disorders is a mess, which we hope (perhaps optimistically) will be improved in the current revisions of both ICD and DSM:

i)	Somatization Disorder The patient has a history of multiple and recurrent medically unexplained symptoms (>6 symptoms) starting in early adult life and lasting for at least 2 years. The symptoms cause distress and impairment and lead to repetitive consultations with medical personnel that are typically unhelpful.  There is usually a history of unnecessary or unhelpful investigations or procedures and the patient may have a high level of disability.  These patients commonly present to many different specialists and are high users of health care resources.

ii)	Hypochondriacal Disorder The patient is persistently preoccupied (for > 6 months) and distressed with the possibility of having one or more serious illnesses. This health anxiety persists despite repeated medical reassurance that they do not suffer from the feared illness(es).  There is overlap with obsessive-compulsive disorder.

iii)	Somatoform Pain Disorder and Chronic Pain The patient has persistent (> 6 months), severe and distressing pain that is not fully explained by a physical disorder and they are pre-occupied by their pain symptoms. Chronic pain is also a common symptom in somatization disorder.

iv)	Conversion (or Dissociative Motor) Disorder The patient has motor or sensory symptoms (e.g., seizures, paralysis, loss of speech, blindness) for which there is inadequate physical explanation. There is usually considerable disability associated with the symptoms.  The patient should not be intentionally feigning the symptoms.  This disorder was of great interest to early neurologists and psychiatrists including Charcot, Janet and Freud, when it was known as hysteria.  The term conversion disorder originally implied that psychological symptoms (or conflicts) were converted to motor symptoms, although this rather simplistic view is now outdated  (Halligan, Bass et al. 2000).  Nevertheless, in practice clinicians treating these patients expect to be able to determine psychological or emotional factors that are contributing to the patient’s presentation.

v)	Body Dysmorphic Disorder The patient has a persistent preoccupation that a part of the body is diseased or deformed, when to an objective observer it is not. The patient will often pursue surgical or other cosmetic treatments in order to correct the perceived deformity and therefore commonly present to dermatologists or cosmetic surgeons.  In ICD-10 this disorder is classified within hypochondriacal disorder, but DSM-IV prefers to keep it as a distinct disorder.  Many psychiatric researchers believe that body dysmorphic disorder would actually be better classified as an anxiety disorder because there is often considerable overlap with obsessive compulsive disorder.

iv) Functional Somatic Syndromes The functional somatic syndromes refer to a number of related syndromes that have been characterised by the reporting of somatic symptoms and resultant disability rather than on the evidence of underlying conventional disease processes. Many such syndromes have been described.  Some of these - such as irritable bowel syndrome - are well recognised within mainstream medicine but others - such as sick building syndrome - are not.  All however share the feature of a disconnection between subjective symptomatology and objective biomedical pathology.  Most medical specialities have at least one functional somatic syndrome – see Table 2 for examples.

Table 2: Functional somatic syndromes by medical speciality

Chronic fatigue syndrome/myalgic encephalomyelitis (CFS/ME), irritable bowel syndrome and fibromyalgia, have been more extensively researched than most other functional somatic syndromes, which has led to specific pathophysiological mechanisms being advanced for each and the development of widely accepted diagnostic criteria. Nevertheless, as yet no specific explanation is compelling and it remains the case that the similarities between the different syndromes are sufficiently striking for there to be a compelling case for considering them together (Barsky and Borus 1999; Wessely, Nimnuan et al. 1999).

Commonly used diagnostic criteria for the three most well known functional somatic syndromes are outlined below:

Chronic fatigue syndrome/Myalgic encephalomyelitis (CFS) (Fukuda, Straus et al. 1994)
 * 6 months disabling fatigue
 * Substantially reduced activity
 * At least 4 of these symptoms:
 * Impaired memory or concentration
 * Sore throat
 * Tender glands
 * Aching/stiff muscles
 * Multiple joint pains
 * New Headaches
 * Unrefreshing sleep
 * Post-exertional fatigue

Irritable bowel syndrome (Rome Foundation, 2006)
 * Recurrent abdominal pain or discomfort at least 3 days per month in the last 3 months associated with ≥2 of the following:
 * Improvement with defecation
 * Onset associated with a change in frequency of stool
 * Onset associated with a change in the form (appearance) or the stool

Fibromyalgia (Wolfe, Smythe et al. 1990)
 * Widespread pain in combination with…
 * Tenderness at ≥11 of 18 specific tender point sites

Epidemiology
Somatic symptoms are common and are the main reason why people seek medical care. Around a third of somatic symptoms that are seen in primary care can be classified as medically unexplained (Kroenke, 2003), whilst the proportion is at least as high in secondary care clinics (Nimnuan, Hotopf et al. 2001; Reid, Wessely et al. 2001; Carson, Best et al. 2003). The prevalence (frequency) of the specific somatoform disorders varies depending on the setting and the diagnostic criteria used. For example the population prevalence of strictly defined somatisation disorder is around 0.5%, but rises to as much as 16.6% when abridged criteria are used (Creed and Barsky, 2004). Likewise the population prevalence of hypochondriacal disorder has been estimated at between 0.02% and 7.7%, with abridged criteria suggesting a prevalence as high as 10.7% (Creed and Barsky, 2004). Body dysmorphic disorder is believed to be present in approximately 1-2% of the general population (Mackley, 2005). Fewer studies have looked at the epidemiology of somatoform pain or conversion disorders, and once again differing diagnostic criteria and populations lead to difficulties in interpretation. Prevalence estimates for the commonest functional somatic syndromes are shown in Table 3. Most epidemiological research in the functional somatic syndromes has focussed on the prevalence of CFS/ME, fibromyalgia and irritable bowel syndrome - probably because operational criteria exist for these disorders.

Table 3: Prevalence figures for a sample of functional somatic syndromes

Most research shows that women suffer from somatoform disorders more frequently than men, with the probable exceptions of hypochondriacal disorder and body dysmorphic disorder. A low level of education is also a risk factor. Other aetiological factors are reviewed below under "Assessment."

Assessment
Allow adequate time Adequate time should be allowed for assessment of patients with medically unexplained symptoms. Although this can be difficult in the setting of a busy primary care clinic or medical outpatients, time spent engaging the patient and gaining a full history will pay dividends later. Patients with severe and enduring medically unexplained symptoms will often have had negative experiences of medical care in the past (Reid, Ewan et al. 1991) (Deale and Wessely, 2001) and it is important that the patient feels believed whenever they are seen by a new health care professional. Therefore good communication skills are important.

Start with the symptoms A good place to start is by taking an exhaustive and full history of all current symptoms. This is not solely for (or even for the purposes of) making a diagnosis, but to demonstrate to the patient that they are being taken seriously and it gives an indication of the way that the patient relates to their symptoms. Duration, severity, exacerbating and relieving factors should be explored for the main symptoms. One of the most neglected questions is to ask the patient what their concerns are about their symptoms (e.g., are they worried that they have cancer?). As a general rule, the more symptoms someone has, the more likely they are to be medically unexplained. It is useful to understand how impaired someone is by their symptoms on a day to day basis and how their illness impacts on their life. When the opportunity arises, psychosocial difficulties should be explored; the easiest way to do this is to use the patient’s own terminology to ask about an area more fully (e.g., if a patient mentions they are "stressed," you can use this word to ask them what is difficult for them in their life). This can help you understand the patient’s illness behaviour better i.e., how does the patient behave when they are symptomatic? Do their symptoms enable them to avoid situations that are stressful? Understanding what the patient attributes their symptoms to can help you explain how unhelpful patterns may have emerged (e.g., a person with CFS/ME who believes their symptoms are due to work stress will behave and manage their symptoms very differently from someone who attributes identical symptoms to a persistent viral infection.

Review previous notes It is preferable to have read previous notes and investigations before meeting the patient, although this is not always possible. It is essential to review old notes before ordering more investigations, as repeating old investigations for previously investigated symptoms can lead to iatrogenic harm (Page and Wessely, 2003). A notes review can add valuable information on previous symptoms or past diagnoses (including somatoform disorders). It also offers an important insight into how the patient interacts with doctors and other doctors’ opinions of the patients’ problems.

Rule out anxiety and depression Patients with anxiety or depression commonly present with physical rather than emotional symptoms. Both anxiety and depression are often experienced physically (e.g., anxiety can present with difficulty swallowing, stomach unease, sweaty palms; depression can present with weight loss, poor appetite, low energy). However, most patients will talk about the emotional symptoms of anxiety and depression if the topic is approached sensitively. Because the terms "anxiety" and "depression" are not universally understood, it is useful to have some probing questions you can use that are suitable for the culture in which you are working. Some examples of questions that are suitable for use in the Western setting are shown in Table 4.

Table 4: Example probe questions when screening for anxiety or depression

Communication For patients with medically unexplained symptoms the first consultation with a new doctor is important. As mentioned above, these patients have often had negative experiences of medical consultations in the past, so an empathic manner and sensitively taken history can be therapeutic in itself. It is never a good idea to imply that you don’t find a patient’s symptoms credible or that there is "nothing wrong" because investigations have been negative. The patient’s symptoms are real and often uncomfortable, even if their patho-physiology is unclear. Many doctors dislike it if a patient expresses negative sentiments about their colleagues or other services. For the most part it is not necessary to enter into an argument with the patient about the rights and wrongs of their previous medical encounters, instead respond to the emotional content of what the patient is saying rather than the specifics (e.g., "that must have made you feel very angry").

Pathogenesis
One issue around all medically unexplained syndromes is when do they become medically explained? Everyone remembers genuine breakthroughs in our understanding of health and disease; one such example being the discovery that General Paresis of the Insane (GPI) (sufferers of which could be found in all the asylums of Europe at the end of the 19th century) was a manifestation of neurosyphilis. When, a generation later, penicillin was found to kill the causative agent, GPI largely disappeared. In our own time, generations of doctors had been taught that peptic ulcer was due to excessive acid secretion, itself the result of stress: that is until Helicobacter Pylori was identified.

But we should also pause for thought. First, the traffic is not all one way. For every previously viewed unexplained or psychiatric illness whose "medical" cause is identified, there is an equal and opposite traffic, as previously viewed medical entities such as visceral proptosis, autointoxication, floating kidneys, chronic appendicitis and so on and so on make the opposite journey. Second, many of the mechanisms that we highlight in this contribution do not cease to be relevant once a causative organism or factor is identified – far from it. The same issues remain relevant, for example psychosocially informed treatments (e.g., Cognitive Behavioural Therapy) do not lose their effectiveness, which is not surprising given that they are of proven efficacy in improving outcome in conditions as diverse as cancer, rheumatoid arthritis, multiple sclerosis, HIV related illness and so on. Somatoform disorders are best thought of as multi-factorial in origin. It is rare than one mechanism (be it emotional or physical) is responsible for a patient’s symptoms. When thinking about why a patient is suffering from medically unexplained symptoms, the traditional psychiatric formulation is helpful i.e., what are the predisposing, precipitating and maintaining factors in this person’s symptoms? It can also be useful to think about how someone’s symptoms may have a physiological (as opposed to patho-physiological) explanation.

Biological Factors
Genetics There is evidence that the general tendency to experience symptoms has a partly heritable basis (Gillespie, Zhu et al. 2000). Furthermore the evidence for the role of genetics in specific somatoform disorders has also increased in recent years. For example twin studies have shown that CFS/ME is substantially heritable (Buchwald, Herrell et al. 2001) and there is also evidence that chronic pain states, including fibromyalgia, might have a genetic component (Buskila, 2007), as might irritable bowel syndrome (Talley, 2006). There may be some genetic liability for hypochondriacal disorder and somatisation disorder, but this has been less investigated (Kendler, Walters et al. 1995; Noyes, Holt et al. 1997).

Neuroendocrine changes Changes within the neuroendocrine system offer an interesting explanation for some of the biological changes seen in the somatoform disorders, although the story is not totally coherent. Most intensive research in this field has been done in CFS/ME and fibromyalgia. There is some evidence of low circulating cortisol in CFS/ME, which is in contrast to the pattern seen in major depression (Parker, Wessely et al. 2001; Cleare 2003). In addition the serotonergic system may be overactive in CFS/ME (Parker, Wessely et al. 2001). A reduction in the responsivity of the hypothalamic-pituitary-adrenal (HPA) axis has also been shown in fibromyalgia (Parker, Wessely et al. 2001). Neuroendocrine changes in irritable bowel syndrome have been less examined, although there is some evidence of abnormal activity of the HPA axis and also that the gut may be over activated by corticotrophin releasing hormone in those with the condition (Fukudo, Nomura et al. 1998). It is likely that at least some of the neuroendocrine abnormalities that have been observed are secondary and these abnormalities are probably best viewed as maintaining factors.

Infection or injury Injury and infection may play a precipitating role in some somatoform disorders and this idea has been most explored for conditions such as CFS/ME and fibromyalgia. In clinical practice patients often cite an injury as the precipitant to chronic pain conditions such as fibromyalgia and this has some limited support in the literature (Al-Allaf, Dunbar et al. 2002). It is generally accepted that there is no single infective agent involved in the pathogenesis of CFS/ME (Afari and Buchwald 2003) or irritable bowel syndrome (Talley and Spiller 2002). Prospective cohort studies - the only way to determine causality - have confirmed that exposure to Epstein-Barr virus (EBV) increases he risk of CFS/ME (White, Thomas et al. 1998), as have Q fever, Lyme Disease (Prins, van der Meer et al. 2006) and viral illnesses requiring hospitalisation (Hotopf, Noah et al. 1996). However, psychiatric morbidity, female gender and prolonged convalescence are still the most important predictors of developing CFS/ME following infection (Hotopf, Noah et al. 1996; Candy, Chalder et al. 2003).

As we write the world’s media are reporting a great breakthrough in the struggle to identify the cause of CFS/ME; a new retrovirus (XMRV) has been identified in 67% of a large series of CFS/ME patients in the USA but only 3% of controls (Lombardi, Ruscetti et al. 2009) - an association that is stronger than that between smoking and lung cancer. The finding is contained in the journal Science, a peer reviewed journal of outstanding reputation. It is indeed genuinely exciting and if true will indeed represent perhaps the single most important change in our understanding of the illness so far. Clinical practice will indeed change, and in the not too distant future, new and novel treatments should emerge. Of course the findings may not stand up to scrutiny, and there have been other equally dramatic claims made in this field before, which have not stood the test of time and replication. But, assuming that this new breakthrough is indeed just that, does that mean that all previous knowledge about CFS is rendered obsolete? Not at all. Perhaps a new drug will abolish CFS, but that seems unlikely. There will remain a major role for the kind of understanding and interventions that are the focus of this chapter, just as they remain important in so many other illnesses and diseases.

Deconditioning Physical deconditioning offers an appealing mechanism for the maintenance of symptoms in the somatoform disorders. There is some evidence for reduced physical fitness in fibromyalgia (Valim, Oliveira et al. 2002) and reduced exercise capacity in CFS/ME when compared to sedentary controls (Fulcher and White, 2000). For patients with chronic and severe somatoform disorders (such as somatisation disorder) the physical effects of years of reduced activity or the use of aids such as wheelchairs can be profound. Such patients present an enormous rehabilitation challenge.

Central dysfunction Some preliminary neuroimaging studies have been conducted in conversion disorder, CFS/ME, irritable bowel syndrome and pain syndromes that suggest that central mechanisms may play a role in these disorders. For example several functional neuroimaging studies have suggested that inhibitory networks are abnormally activated in conversion disorder (Aybek, Kanaan et al. 2008). At present, the usefulness of neuroimaging research in somatoform disorders is limited, but taken as a whole probably does support the idea of aberrant patterns of brain activation in these conditions (particularly in response to relevant probes such as experimentally induced pain). It is not known whether these changes pre-exist the illness or have developed secondarily.

Psychological Factors
Childhood experiences Experience in childhood appears to be relevant to the development of somatoform disorders later in life. Longitudinal studies show that children who experience parental ill health in childhood are more likely to develop medically unexplained symptoms as adults (Hotopf, Mayou et al. 1999). Whether childhood illness increases the likelihood of adult somatoform disorders is less clear – certainly childhood medically unexplained illness appears to do so (Hotopf, Wilson-Jones et al. 2000). Childhood sexual abuse increases the risks of adult somatoform disorders (Paras, Hassan Murad et al. 2009).

Stressful events Stressful events can precipitate the onset of a somatoform disorder and are known to occur more frequently in the period leading up to the onset of medically unexplained symptoms (Craig, Drake et al. 1994). A similar picture has been shown for CFS/ME, with patients experiencing "dilemmas" in the months preceding onset (Hatcher and House, 2003). Chronic stress (or life events) has also been shown to be important in the onset and maintenance of symptoms in irritable bowel syndrome (Creed, Craig et al. 1988; Bennett, Tennant et al. 1998) and fibromyalgia (Anderberg, Marteinsdottir et al. 2000). Trauma such as sexual abuse (Paras, Hassan Murad et al. 2009) or involvement in a disaster (van den Bergh, Grievink et al. 2005) also appears to be a risk factor for the development of a range of somatoform disorders

Personality It is often presumed that personality factors are an important predisposing and maintaining factor in somatoform disorders, although there is little supporting evidence. Emotional instability (or neuroticism) may prolong the course of hypochondriacal disorder (olde Hartman, Borghuis et al. 2009) and, along with introversion, has been found to be a risk factor for the development of CFS/ME (Kato, Sulllivan et al. 2006; Prins, van der Meer et al. 2006). Patients with non-epileptic seizures (a type of conversion disorder) have been found to have high rates of personality disorder (Bowman and Markand, 1996). Clinically, patients with co-morbid personality disorder can be very challenging to manage.

Illness beliefs Illness beliefs are enormously important in the maintenance (and possibly precipitation) of somatoform disorders. Beliefs link bi-directly to both behaviours and emotions, which means that by altering one of these domains the other two are likely to be affected – see the diagram below. Patients with CFS/ME are more likely to make physical illness attributions for a selection of common symptoms compared to controls (Butler, Chalder et al. 2001); perhaps in consequence and are more likely to believe their illness will be chronic and have serious consequences when compared to patients with chronic medical conditions (Weinman, Petrie et al. 1996). Illness worry is related to disability in fibromyalgia, but not in rheumatoid arthritis (Robbins and Kirmayer, 1990). Likewise, those with irritable bowel syndrome score more highly on hypochondriacal and bodily preoccupation scales than control groups (Gomborone, Dewsnap et al. 1995).

Making physical attributions for unexplained symptoms is natural – the problem is what these may imply for the person’s concepts of self efficacy, acceptable treatment and likely prognosis. Deale et al showed that for patients with CFS to recover, it was not necessary that their illness attributions changed (e.g., "the illness is physical and caused by a virus"), but instead improvement was linked to change in beliefs such as "doing too much makes me worse," "I need to rest to get better" and so on (Deale, Chalder et al. 1998). In other words, physical illness attributions can act as a confounder or marker for more unhelpful beliefs that are associated with maladapative coping responses. The term "symptom amplification" is used to describe the manner in which innocuous symptoms become incorrectly attributed and then incorporated into a patient’s understanding of their illness, which leads to further incorrect attribution of other symptoms as they arise (Barsky and Borus, 1999). These beliefs and attitudes about symptoms may act as a mechanism that then guides the patient to adopt avoidant behaviours, leading to limitation of activity, which in turn leads to the secondary deconditioning and neuroendocrine effects outlined above. Avoidance behaviours are invariably based on the patient’s understanding of their illness (e.g., "when I feel fatigued, I cause myself further harm if I exercise") and are often possible to work with during treatment (Deale, Chalder et al. 1998).

Socio-cultural Factors
Living Environment We have discussed above the possible importance of the early family environment in predisposing someone to suffer from medically unexplained symptoms (e.g., by experiencing the illness of a parent in early life), however the behaviour and attitude of close family and friends can also play a role in the maintenance of medically unexplained symptoms. For example partners of patients with CFS/ME are more likely to make physical attributions about their partner’s symptoms than the partners of fracture clinic patients (Butler, Chalder et al. 2001). Clinically it is often as relevant to understand the illness beliefs of close family as it is the patient’s – particularly when the family are providing high levels of care and support.

Financial Reward The financial "reward" to be gained from disability payments or litigation has been argued to play a role in the maintenance of ill-health in those suffering from somatoform disorders (Malleson, 2002). For example, being in receipt of sickness benefit or certification has been shown to be a poor prognostic sign in CFS/ME (Cope, David et al. 1994; Bentall, Powell et al. 2002) and fibromyalgia (Wigers, 1996), whilst the whiplash syndrome does not appear to exist in countries without an insurance/compensation culture (Schrader, Obelieniene et al. 1996).

Media Several functional somatic syndromes including CFS/ME, Gulf War syndrome and repetitive strain injury, have gained public credibility in spite of widespread medical scepticism as to their very existence. The role of the media in this process has often been highlighted (Shorter, 1995; Barsky and Borus, 1999; Hazemeijer and Rasker, 2003). The availability and explosion in internet sites has also meant that patients may inadvertently be exposed to information that is inaccurate or even harmful (Armstrong, 2000; Kisely, 2002).

Treatment
The assessment itself can be therapeutic, particularly if time is taken to provide a clear explanation for symptoms, which is not perceived by the patient to blame them. The doctor may need to avoid colluding with the patient, but also avoid denying the reality of the symptoms. Research has shown that "empowering" explanations are the most beneficial for patients with medically unexplained symptoms (i.e., explanations that provide a tangible mechanism, de-emphasise blame and provide the opportunity for self-management) (Salmon, Peters et al. 1999). The provision of clear information in different forms (i.e., verbal and written) is necessary. Patients with medically unexplained symptoms often appear to be seeking reassurance, but this can be difficult to deliver effectively. It is counter-productive to tell a patient that "there is nothing wrong," when their symptoms are proof that there is. On the other hand it is important to counter specific illness fears that the patient may hold (e.g., "My symptoms mean I’ve got cancer," "This rash shows that I have HIV," "If I do too much I will permanently damage my spine") if that is not the case. This is why it is important to have asked the patient what they believe is wrong. Patients with hypochondriacal disorder will often attempt to elicit repeated reassurance, which fails to provide reassure for any length of time (Deale, 2007).

If there is evidence of anxiety or depression at first assessment, then this should be treated in the usual way. Doing so will often, although not always, lead to a significant improvement in the patient’s somatic symptoms. A doctor that sees a patient with medically unexplained symptoms for follow-up has an important role to play in managing that patient’s interaction with medical services. Even if the doctor does not perceive themselves to be providing active therapy, they can be aware of potentially iatrogenic interventions (i.e., harm caused by doctors) (Page and Wessely, 2003). They can also provide regular follow-up that is not contingent on the patient being symptomatic, thereby discouraging the need for the patient to complain of symptoms in order to elicit care. It is sometimes possible to agree beforehand that only a certain proportion of the session will be devoted to discussing symptoms, and leave it to the patient to decide the content of the second half of the interview.

Psychotherapy
Overall cognitive behavioural therapy (CBT) is known to be an efficacious treatment for the range of the conditions loosely grouped under the somatoform disorders (Sumathipala, 2007). CBT and similar therapies have shown specific usefulness in the treatment of hypochondriacal disorder (Thomson and Page, 2007), CFS/ME (Chambers, Bagnall et al. 2006), irritable bowel syndrome (Brandt, Bjorkman et al. 2002), fibromyalgia (Rossy, Buckelew et al. 1999) and burning mouth syndrome (Zakrzewska, Glenny et al.). CBT can be adapted for use in any of these disorders, but like most medical treatments relies on the patient being sufficiently motivated to participate. One of the first goals in CBT is for the therapist and patient to come to a shared understanding of the patient’s problems using a CBT framework – the therapist often uses diagrams like the one on page 18 to illustrate this. Evidence is lacking for useful psychotherapeutic treatments for conversion disorder (Martlew, Baker et al. 2007), although preliminary studies have shown that, once again, CBT may be useful (Goldstein, Deale et al. 2004).

Pharmacotherapy
Overall there is evidence that antidepressant medication is useful in the treatment of somatoform disorders (O'Malley, Jackson et al. 1999; Sumathipala, 2007), although it is not possible to generally recommend the use of one type of antidepressant over another. For the functional somatic syndromes there are some specific recommendations, for example tricyclic antidepressants are effective in treating fibromyalgia (Arnold, Keck et al. 2000), abdominal pain in irritable bowel syndrome (Brandt, Bjorkman et al. 2002) and premenstrual tension (Steiner, Steinberg et al. 1996). On the other hand antidepressants have not been found to be useful in CFS/ME without co-morbid depression (Whiting, Bagnall et al. 2001). In general the effectiveness of antidepressants in these disorders increases if the patient has evidence of co-morbid depression or anxiety, however medication is probably less effective than psychological approaches.

It can be necessary to rationalise inappropriate medication, as some patients with somatoform disorders are prescribed medication that is unnecessary or even harmful. This needs to be done by (or in conjunction with) primary care and the rationale discussed with the patient in advance.

Combined treatments
In clinical practice it is common to combine a psychotherapeutic and pharmacological approach to management. The patient may have strong feelings about treatment and these should be taken into consideration. In developed countries treatment for somatoform disorders can sometimes be provided by specialists (e.g., consultation-liaison psychiatrists) attached to general hospitals, although provision is often patchy and as in developing countries much of the burden falls to primary care.

Final Considerations
Factitious disorder or Munchausen’s syndrome is listed separately (adjacent to the somatoform disorders) in DSM-IV classification, whilst ICD-10 classes it amongst the personality disorders. Malingering is not considered to be a psychiatric disorder by either system. However the distinction of factitious disorder or malingering from the somatoform disorders can be unclear, so for the sake of completeness we mention them here. Diagnostic features are outlined below. Factitious disorder is probably a rare condition about which little is known, although persons suffering from this disorder are likely to have significant personality disturbance and a background of neglect or abuse. Malingering is more common, although quite how common is unknown due to the nature of the behaviour.

Factitious disorder
•	Persistent faking of symptoms or self-infliction of wounds to produce symptoms •	Persistent visits to hospital in order to gain care for these symptoms (may move from hospital to hospital to avoid detection) •	No external gain (e.g., financial) is apparent, so the gain is viewed as being psychological

Malingering
•	Deliberate falsification of a medical condition •	The falsification (or exaggeration) is for financial or other obvious material gain