The Psychologisation of Illness

Ellen Goudsmit


Psychologisation describes the emphasis on psychological factors where there is little or no evidence to justify it (1). It's a process where relevant findings are ignored or downplayed in favour of data from incomplete examinations, flawed research or anecdotal reports. In a clinical context, differential diagnoses may be dismissed prematurely while psychological explanations are readily accepted. Psychologisation does not refer to situations where there is sound evidence that psychological factors play a significant role, or where all the arguments are discussed and the psychological explanations are deemed the most persuasive.

Psychologisation is a serious issue because it leads to misdiagnosis, inappropriate treatment and unnecessary psychological distress (2). Moreover, it undermines the general population's confidence in orthodox medicine and reduces their trust in its practitioners. Yet, despite the dangers, psychologisation has received little attention in medical texts. In this article, I will discuss the main reasons underlying psychologisation and examine whether it has influenced views about food allergies and intolerance.

Psychologisation in clinical practice.

Many cases of psychologisation in clinical practice occur because doctors jump to conclusions without checking that their psychological explanations are correct. For instance, Treasure et al (3) described how three women suffering from "breathlessness and panic symptoms" were admitted to their accident and emergency department. The initial diagnosis was hysterical hyperventilation, as a result of which two of the patients were referred to a psychiatrist while the third was discharged with a prescription for diazepam. Within three days however, the women returned and tests revealed that they were suffering from diabetes. According to Treasure et al, the assumption that the problem was psychological had led "to the dangerous omission of physical examination, basic nursing observations, and urine analysis. In all these patients a simple test for glycosuria would have made the diagnosis obvious and the consequent and considerable risk to the patients could have been avoided."

Another typical example is that of a woman who developed anxiety, fatigue and intermittent cramp-like abdominal pains after changing her job. Her GP diagnosed irritable bowel syndrome (IBS), precipitated by the stress at work, and advised a short period of sick leave. As the pain did not subside, she continued to consult her GP but neither he, nor the other doctors who saw her, questioned the diagnosis. A few months later, she noticed that the pain was associated with "spasms visible through the abdominal wall", but the doctors who examined her at the local casualty department continued to blame IBS. When the patient consulted the author of the report, she found "an easily palpable mass in the left iliac fossa" and an emergency operation confirmed cancer of the sigmoid colon. As in the case above, doctors had overlooked organic disease and wrongly attributed all the symptoms to a psychological cause.

Psychologisation in the literature on aetiology.

Some of the most interesting examples of psychologisation can be found in the theoretical discussions of disease. For instance, Booth (5) offered a psychodynamic view of tonsillitis, arguing that many cases represented a love relationship with a bacteria in which the patient establishes "libidinous contact on the earlier oral instead of the later genital level... the object of the oral strivings is no longer another human being, but bacteria... Such substitutions of bacteria for a human being makes biologic sense considering the fact that the basic object of the sex act is the copulation of the sex cells: primitive unicellular organisms like bacteria."

This is a typical case of psychologisation, not only because of the poor quality of the supportive evidence but also because the author ignored all the relevant data to the contrary. For more recent examples of speculative and over simplistic hypotheses, see Hay and Millenson (6, 7).

Most cases of psychologisation are not as extreme and indeed, may not be recognised except by colleagues who are familiar with the research. Perhaps the best examples of the more subtle forms of psychologisation can be found in the British medical literature on myalgic encephalomyelitis (ME) and chronic fatigue syndrome (CFS)1. Many articles on these conditions feature highly selective discussions of the scientific data, with an emphasis on findings relating to psychiatric morbidity and a tendency to ignore or dismiss evidence of disease (8).

The problems began in 1970, with the publication of a paper on the epidemic, which closed the Royal Free Hospital during the summer of 1955. 'Royal Free disease' resembled a viral infection and affected over 200 members of staff, the majority of whom were female. It was eventually reclassified as ME but one of the consultants involved in their treatment noted later that he and his colleagues had considered the possibility of hysteria. However, "the occurrence of fever in 89%, of lymphadenopathy in 79%, of ocular palsy in 43% and of facial palsy in 19% rendered it quite untenable"(9).

Fifteen years after the event, psychiatrists McEvedy and Beard were given access to some of the incomplete reports as well as the notes of those who were not thought to have had 'Royal Free disease' (10). However, instead of restricting their conclusions to the sample studied, they dismissed the whole outbreak as a case of mass hysteria.

They based their view on three main arguments. Firstly, they claimed that most of those affected had been young and socially segregated young females. Secondly, they argued that the illness had not spread beyond the institutional population. Thirdly, they alleged that it had not affected a significant number of men. Notable omissions in their account included what might have triggered the anxiety among the older and more experienced health professionals involved, or what may have led to the lymphadenopathy or the morphological abnormalities seen in circulating lymphocytes (11, 12).

In fact, there are many arguments against the mass hysteria theory (9, 13, 14), although these are generally ignored (e.g. 15, 16). For example, the nurses were not socially segregated as suggested by McEvedy and Beard (17) and in terms of age, the illness was more common amongst those over forty (11). As for the predominance of females, this was true only in relation to the non-resident staff. The attack rates for resident men and women were 19 and 20 per 100 respectively (11).

It has also been pointed out that if the epidemic had resulted from anxiety in a segregated group of women, then one would have expected a higher proportion of cases in the Elizabeth Garrett Anderson Hospital and Maternity Home (18). These institutions were part of the Royal Free Hospital group and were run for and staffed entirely by women. However, less than 7% of the patients came from these units - and only the Eastman Dental Hospital recorded fewer cases (11).

Another argument against McEvedy and Beard's explanation is that some of the features of the outbreak were simply not typical of mass hysteria. According to Sirois (19), most of these episodes last between 10 and 20 days and primarily affect adolescent women. In contrast, the outbreak at the Royal Free Hospital lasted for four months (from July 13th 1955 until November 24th) and only a minority of those affected were younger than 20 (11).

Finally, the literature on sporadic cases, which clearly resembled those seen in the epidemic, proved that the outbreak not only spread beyond the institutional population (20, 21, 22) but also beyond London (9). Moreover, there was evidence of viral involvement both in the Royal Free patients (Parish 1994, personal communication) and in people with an identical disorder examined later (22). On the basis of these findings alone, the suggestion that all the cases at the Royal Free were the result of mass hysteria is unsustainable.

In the light of the available evidence, McEvedy and Beard's article should have been dismissed as deeply flawed and biased. Instead, it was widely accepted and became highly influential. For many years, their conclusions diverted the scientific community's attention away from the virological and neurological aspects of the disease and thus limited the research, and our understanding of ME. More importantly, their dualistic 'mind or body' approach provided the basis for the organic versus psychological debate, which has dominated the literature on the illness for the past two decades (15, 23, 24).

The causes of psychologisation

1. Stereotypical views of women

Although the literature shows that men are not immune, most of the reports relating to psychologisation feature women. In older texts, female patients were often portrayed as suggestible, emotionally unbalanced, irrational, manipulative and unable to cope with relatively minor 'stress'. For instance, authors writing about dysmenorrhoea claimed it was more common in "highly strung", "nervous" or "neurotic" women, and speculated how a 'faulty outlook' might lead to "an exaggeration of minor discomfort" (25).

Premenstrual syndrome was also regarded as a mild and benign problem which women should be able to endure without medical intervention. As in the accounts of dysmenorrhoea, authors often implied a link with somatisation, with suggestions that certain women were guilty of exaggerating their distress, perhaps to escape responsibility or effort (26). Conversely, most of these articles did not consider the effect of recurring symptoms on the women's self-confidence and self-esteem, or the possibility that this could explain the raised scores on psychometric tests (27). Ignoring virtually all arguments to the contrary, many authors regarded emotional problems as the cause, rather than the result of ongoing ill-health (25).

A more specific illustration of the influence of gender stereotypes was reported by Roberts (28). She related the story of a brother and sister, both of whom suffered from migraines (28). When the doctor was asked about his thoughts on the woman, he attributed her headaches to her social situation - tensions and problems caused by the fact that she was not married, did not have children, and that she was a 'career woman'. In contrast, he regarded her twin brother's migraines as having a genetic and biochemical basis and refused to consider a social explanation.

One reason why some doctors may be more inclined to psychologise the illnesses reported by women is the way in which many express their distress. Discussing her experiences as a physician, Gillespie (29) suggested that: "men tend to speak from fact, women tend to speak from emotion." Thus, women may refer to pain in terms of ruining their marriage and destroying their life. They relate the effects of their symptoms rather than their nature. According to Gillespie, this emphasis on the mental anguish can mislead some doctors into assuming a psychological aetiology.

A similar phenomenon has been documented in relation to ethnic groups. In one study, Zola (30) found that Italian Americans were given more psychiatric diagnoses than patients with an Anglo-Saxon or Irish background. Accounting for the findings, Zola noted that Italians had a tendency to report the effects of their symptoms and he suggested that their behaviour could have been interpreted as "overacting". As in the case of women, the way in which these patients described their symptoms might have influenced the clinicians to focus on one explanation and to dismiss another.

Some of the stereotypical ideas about female patients may have originated during medical training. As one female student told Howell (31): "women's illnesses are assumed psychosomatic until proven otherwise." Another observed that women are often "portrayed as hysterical or as nagging mothers or as having trivial complaints. Men are almost never pointed to as having a psychological component to their illnesses - this is generally attributed to women."  Howell also reported that mothers were sometimes described as "complaining", and that older women were deemed to be "demanding and bitchy".

The belief that psychological factors play a major role in the conditions reported by women may be one reason why they are often investigated less thoroughly than the same symptoms in men. For example, a small study by Armitage et al (32) revealed that physicians ordered more extensive tests and procedures for men than for women with the same complaint.  Discussing the findings, the authors suggested that the physicians "might have been responding to current stereotypes that regard the male as typically stoic and the female as typically hypochondriacal".

Such a view is supported by Ayanian and Epstein (33) who studied patients hospitalised for coronary heart disease and found that the women underwent fewer diagnostic and therapeutic procedures than the men. Likewise, Steingart et al reported that women with cardiac disease were offered fewer procedures like coronary by-pass surgery than men, even though the women were more disabled and those procedures could have significantly lessened the symptoms (34). The authors stated that the symptoms of chest pain in women were more likely to be attributed to non-cardiac causes but they did not speculate which. In a more recent study, Pope et al (35) assessed patients with acute cardiac ischemia who were mistakenly discharged from the emergency departments of ten American hospitals. They found that discharge was more likely if the patient was a woman aged 55 years or younger, 'non-white', or if the principal symptom was shortness of breath. Although this study does not illustrate overt psychologisation, it shows that in this age of evidence-based medicine, variables like gender, age and race continue to influence some doctors' clinical judgement.

Despite the large number of reports detailing apparent 'sex-discrimination', the extent of gender bias in medicine is difficult to assess. For instance, several analogue studies found no significant differences in the percentage of men and women who were given a psychogenic diagnosis or psychotropic drugs (36). However, it is probably fair to say that stereotypical attitudes still play a role in many cases of psychologisation, though to varying degrees and alongside other variables such as ethnic background and age.

2. Lack of clarity relating to symptoms.

Another important factor underlying many cases of psychologisation is the difficulty in diagnosing certain conditions. For instance, Himmelhoch et al (37) described eight patients with subacute encephalitis, of whom seven were initially diagnosed as having functional psychiatric disorders. They concluded that: "the bizarre behaviour induced physicians to ignore neurological findings, to overlook evidence of organic syndromes (such as intermittent lucidity and markedly abnormal electroencephalograms) and to make functional diagnoses."

Likewise, Randy Shilts noted in his history of the AIDS epidemic that: "doctors frequently missed the damage to the central nervous system, writing off the often vague symptoms of dementia as related to stress or depression" (38). Similarly, the predominance of non-specific symptoms like weakness and fatigue have been blamed for the failure to correctly diagnose cases of myasthenia gravis (39).

3. Lack of evidence.

The lack of clarity in relation to symptoms is not the only factor which can complicate the diagnostic process. There are now so many laboratory tests available, that some doctors may interpret 'absence of evidence' as 'evidence of absence' (40). Thus if tests fail to identify a physical cause, some may reason that there is no physical cause to be identified and that the symptoms therefore have a non-organic, i.e. psychological cause. Consequently, patients may be told that "there is nothing wrong" or that the illness "is all in your head"; two assessments which not only contradict actual experience but which also imply that the distress is not legitimate. Where this occurs, the physician's failure to define the illness as a 'disease state' is often an additional source of suffering which can undermine the patient's psychological health. As Stewart and Sullivan (41) observed, the emotional conflicts and tension caused by the 'misdiagnosis' of pathology can lead to "a type of iatrogenic disease" characterised by symptoms such as "feelings of frustration, worry and intermittent periods of depression."

One person to have experienced this type of psychologisation is the late Jacqueline Du Pré. In her biography of the cellist, Carol Easton (42) wrote:

"There is no specific test for multiple sclerosis.  Its early symptoms - fatigue, loss of sensation, weakness and visual changes - are frequently misdiagnosed as psychoneurosis or an even more severe psychiatric disorder, such as hysteria, particularly in women. When doctors could find no organic cause for her complaints, they prescribed a year's rest, and referred her to a psychiatrist... When she consulted a doctor in Australia about her tenacious fatigue and occasional double vision in her right eye, he dismissed her symptoms as "adolescent trauma" and suggested she take up a relaxing hobby."

According to the Multiple Sclerosis Society (personal communication), such an experience is not rare and after months or years of having been told their symptoms are psychological or psychosomatic, many patients feel relieved when they learn that they have MS. Robinson (43) confirmed this in his study, stating that: "the disclosure of the diagnosis gave back to many respondents their credibility and legitimised their strange behaviours which had previously been labelled as neurotic, hypochondria, malingering or drunk."

Further evidence that diagnostic uncertainty can influence attitudes towards patients comes from a study conducted among nurses (44). They were asked to rate the degree of suffering, the need for pain relief and the personality of patients with severe pain. One of the vignettes they were given listed signs of objective pathology, the other did not. The results showed that the nurses attributed less intense pain when the patient in the vignette had no signs of pathology or when the condition was chronic. Moreover, ratings of the patient's personality were significantly more positive when pathology was present than when it was negative. According to the researchers, the nurses relied on the biomedical model of pain which suggests that this symptom is a physical response to tissue damage. The failure to find evidence of tissue injury led to a psychogenic labelling, as did chronicity, the suggestion being that after a time patients should have adapted to their pain.

4.  Cause or effect?

Some cases of psychologisation may be due to the reversal of cause and effect. For instance, in their review of articles on severe dysmenorrhoea, Lennane and Lennane (25) found that doctors tended to view their patients' fear and dislike of menstruation as a cause of their pain, rather than as a result. Similarly, the frustration and irritability documented in women with mastalgia were for many years dismissed as an expression of psychoneurosis, rather than the effects of recurrent pain. However, when researchers studied these women, they were found to be "no more neurotic" than people with varicose veins (45). It is now generally accepted that most cases of mastalgia are related to hormonal imbalance (Brush, personal communication).

The tendency to view psychological states and traits as a cause rather than the result of illness might also explain some of the psychological explanations for Parkinson's Disease (46) premenstrual syndrome (27), cancer (47), multiple sclerosis (48, 49) and the diabetes personality (50).

5.  Interpretation of research findings.

When research reveals an apparent association between illness on the one hand and psychological symptoms, states and traits on the other, it is often difficult to determine the direction of the relationship. In addition to the possibility that the psychological variable caused the disease, there are a number of other explanations which should be considered (51). They include:

a. Coincidence. The likelihood that both the illness and the psychological variable may have been caused by independent agents and that they are not related.

b. The presence of an external agent.  A third variable, for instance, environmental pollution, might have caused both phenomena.

c. The existence of an internal agent. For example, heredity or an immunological or neurological dysfunction may have influenced both psychological and physical variables.

d. The influence of a behavioural agent. Certain behaviours may have induced emotional states and affected the person's health. For instance, eating a poor diet could result in depressed mood as well as disease. 

e. The presence of an intervening agent. Here an event or state intervenes between the psychological and biological factors, causing them to correlate. For example, depression caused by bereavement may compromise immune function which then increases one's predisposition to or exacerbates ill-health. 

f. Disease causation. A sixth possibility is that a disease caused a particular psychological reaction. Since it is often difficult to pinpoint the actual onset of a disease, it is possible that certain physiological changes preceded the psychological ones. Thus depression may be the first symptom of pancreatic cancer, multiple sclerosis or Parkinson's disease (52).

The interpretation of an apparent link between illness and psychological variables is further complicated by methodological problems, notably the use of different diagnostic criteria and measures. This has resulted in an astonishing range of prevalence figures for psychiatric morbidity in people with cancer, diabetes, multiple sclerosis and CFS (8, 53, 54, 55). Examples of psychologisation include the citation of higher estimates in support of a 'psychosomatic' explanation, and not documenting, or unfairly dismissing lower rates. In other cases, authors may deliberately overlook the flaws associated with a particular measure, leading them to exaggerate the significance of psychiatric illness associated with a disease.

For instance, it is well known that estimates may be inflated as a result of confounding, i.e. the inclusion of symptoms of the medical condition as criteria for the diagnosis of psychiatric disorders (56, 57, 58, 59). The symptoms most often used for this purpose include fatigue, insomnia, loss of appetite, psychomotor retardation and difficulties with concentration (60, 61, 62, 63, 64, 65). Not surprisingly, omitting one or more of these from the list of criteria can have a significant effect on the estimates of psychopathology. For example, in their study on chronic fatigue, Katon et al eliminated fatigue as a criterium of depression and found that this alone reduced the prevalence rate from 15.3% to 10.2% (66).

The inclusion of disability-related items in self-rating scales can cause similar problems. For instance, Yeomans and Conway (53) reported that 33% of their patients with ME scored 11 or more on the depression subscale of the Hospital Anxiety and Depression questionnaire (HAD). However, when they excluded the item 'I feel as if I am slowed down', no one exceeded the cut-off point for caseness.

The reliance on symptoms can also lead to false positives in relation to somatisation disorder (SD). For instance, estimates of SD in patients with CFS have varied from 0% to 98% (67). When one team of researchers excluded 7 somatic items from the diagnostic criteria, this reduced the prevalence rate from 46% to 20% (66). Reduced rates have also been noted by others, leading Johnson et al (67) to conclude that the "diagnosis of SD is of limited use in populations in which the aetiology of the illness has not been established."

In order to assess whether the psychological disturbances reported by people with CFS represent a reaction to their illness or the presence of a primary psychiatric condition, researchers have compared this population with people suffering from multiple sclerosis (68, 69), rheumatoid arthritis (66), cystic fibrosis (70), spinal cord injuries (8) and muscle disorders (71, 72). Unfortunately, it is difficult to interpret most of these studies because they failed to match the samples with regard to the degree of impairment or the severity of symptoms. In relation to fatigue, the levels recorded by the comparison groups were often significantly lower than those of the patients with CFS. Where fatigue was not assessed, other illness-related variables showed that the controls had fewer symptoms than the people with chronic fatigue (66).

Despite the flaws, most reviewers have concluded that patients with CFS have higher rates of psychiatric disorders than other disabled groups. However, the failure to control for the degree of impairment means that their emphasis on psychopathology as a major factor in the perpetuation of CFS is probably not justified (73, 74, 75).

6. Psychologisation and the illusion of control.

While psychologisation can have serious and occasionally fatal consequences (40, 76, 77), it also offers certain advantages (78, 79). For instance, Taylor (79) found that many women with breast cancer believed that stress management techniques and a positive attitude would prevent a recurrence. Although there is little research evidence which supports this view (80, 81), both she and Turnquist et al (82) consider these 'illusions' to be adaptive, enabling more women to cope with the fear, vulnerability and helplessness associated with their illness.

The reverse is true when outsiders claim that patients create and are responsible for their symptoms (6, 10). A discussion of victim-blaming statements can be found in texts by Sontag (83) and Wilkinson and Kitzinger (84). For instance, Sontag documents how cancer patients were once described as having a "great tendency for self-pity" and as being "empty of feeling and devoid of self". In more general terms, it was alleged that the "sick man himself creates his disease" and that "illness is in part what the world has done to a victim, but in a larger part it is what the victim has done with his world, and with himself."

In her response, Sontag concludes that "such preposterous and dangerous views manage to put the onus of the disease on the patient... The view of cancer as a disease of the failure of expressiveness condemns the cancer patient; expresses pity but also conveys contempt." To that might be added that the emphasis on the spiritual flaws of the sick and disabled will make many patients feel guilty, thus further increasing their distress.

So who gains from this approach? Outsiders might be tempted to play down the influence of factors such as viruses and pollution - over which mankind has relatively little control - to enhance their 'sense of mastery' over their lives (79). By attributing them to personal flaws (which one doesn't share), one reduces one's own perceived risk of being struck down by a feared disease.

A variation on this theme is the suggestion by Lerner and Simmons (78) that 'blaming the victim' helps people to maintain their belief in a 'just world', where people get what they deserve and deserve what they get. According to this hypothesis, individuals can avoid adversity through their own activities but this also implies that those struck by misfortune are deemed to have somehow merited their fate.

Surprisingly, such a view of illness is not restricted to a small group of 'New Age' philosophers or the odd clinical text (6). Wortman and Dunkel-Schetter (85) wrote that health care professionals "often have ambivalent feelings towards their patients - patients they are supposedly trying to help. Past research has suggested that even a single encounter with a victim can be a powerfully distressing experience, and can result in blame and derogation of the victim." Thus the occasional comment which insinuates that the patient's character made them ill may be part of a coping strategy which allows practitioners to protect themselves from unpleasant emotions, even though it will almost certainly undermine the relationship with those in their care.

Finally, blaming the victim may result from strongly held political views (84). For instance, some believe that health is a matter of personal responsibility and that people increase their risk of disease by engaging in behaviours like smoking or eating an unhealthy diet. The politically motivated may exaggerate the influence of lifestyle or apply this reasoning to other conditions, to limit funds destined for research and restrict patients' eligibility for certain benefits. Thus they will argue that if the cause of illness lies in people's character and behaviours, then fewer resources have to be spent evaluating the effects of environmental pollutants, pesticides and other external factors implicated in disease. Unless governments redirect the savings to educational projects concerned with prevention, or to support those already affected, blaming the victim is a useful way of freeing-up funds and diverting resources.

Psychologisation of food allergy and intolerance.

Examples of psychologisation can also be found in the literature on food sensitivities. Most of these focus on the apparently high prevalence of psychiatric disorders in this population, with the implication that these are the main source of the patients' complaints (86). However, as will be argued in the following section, such suggestions should be interpreted with care.

A major difficulty in assessing the actual role of psychological factors in food allergy and intolerance is that the majority of the studies have included patients with multiple chemical sensitivities and 'environmental illness' (87, 88). To complicate matters, some of the subjects were engaged in legal action for compensation at the time (89), and many had supposed rather than proven allergies (90). Indeed, few studies have restricted participation to patients with properly diagnosed food sensitivities and even fewer have assessed delayed reactions to food, presenting 24 hours or more after ingestion.

Another problem is the assessment of psychiatric morbidity and the use of measures which include the symptoms of food sensitivity as criteria of psychopathology. As noted above, such confounding can lead to diagnostic false positives (87, 88) and give a misleading view of these patients' mental health.

This is certainly the case in relation to somatisation disorder, where the criteria in DSM-IV require the presence of seven symptoms from a list which includes nausea, bloating and intolerance to several different foods, as well as headache and abdominal pain (56). Accordingly, it is advisable to obtain additional evidence such as the presence of conflicts and other stressors at meal times or obvious signs of secondary gain, before attributing these symptoms to SD. If the aetiology is unclear, it may also be useful to engage the patient in an experiment, for instance, to compare the effects of standard treatments such as avoidance with the results of psychotropic medication (91). A positive response to the latter does not prove that the reaction to food was mediated by emotional as opposed to immunological factors, but it could give useful insights into the mechanisms underlying specific symptoms, and clarify the role of contributory factors.

Some clinicians consider the documented rates of psychopathology to be reliable, arguing that symptoms such as depression are generally not triggered by food (90). However, others remain unsure and have pointed to evidence linking diet with a number of psychological disturbances, including affective disorders and schizophrenia (7, 92, 93, 94).

As noted above, the finding of high rates of psychopathology does not prove that those disorders caused the patients' symptoms. Depression and anxiety may be due, at least in part, to factors such as uncertainty. In previous research, uncertainty relating to conditions like myocardial infarction, multiple sclerosis and CFS was significantly correlated with psychological distress (8, 95, 96). It is therefore possible that this variable also plays a role in people with food allergies and sensitivities. Similarly, any mood disturbances may be exacerbated and perpetuated as a result of continued psychologisation by physicians and family (41, 97, 98). These are just two of the factors which should be considered when assessing the psychological status of individuals with food sensitivities (8). Focusing exclusively on the prevalence of affective disorders could lead to mistakes.

In differentiating cause and effect, a prior history of psychiatric illness is often regarded as suggestive of a psychological vulnerability (99). The problem here is obvious. Without corroboration from accurate medical notes, much depends on the patient's memory and their own interpretation of their complaints (88). Moreover, if one has to rely on the patient, it must be remembered that current psychological distress can increase the tendency to recall some symptoms and health events at the expense of others. Moreover, it is possible that past symptoms were the result of conditions like malabsorption and altered gut ecology, which might have predisposed patients to develop sensitivities to food later in life (100). Leaving these considerations aside, there is at present little evidence that people with proven food reactions are more likely to have a history of psychiatric morbidity or that they have a greater psychological vulnerability than people who do not experience reactions to foods (101).

Finally, it has been suggested that blaming symptoms on allergies and environmental toxins may help people with psychiatric disorders to avoid the stigma of mental illness (24, 86). In a society where the general population and certain physicians still regard psychiatric disorders as less worthy of attention and respect than physical ones, it is understandable that some patients might choose to blame their unexplained symptoms on allergies rather than depression or anxiety. One can also postulate that these erroneous attributions may be reinforced by some self-help groups and articles in the popular press (102). However, there is still no convincing evidence that most patients with food allergies are motivated by prejudiced views of mental illness. Moreover, until every patient is also assessed for conditions like dysbiosis and tested for immune activation (103), one cannot assume that their attributions are incorrect.

In summary, most of the claims that psychological factors are the source of food sensitivities seem to be based on methodologically flawed research and a biased interpretation of the findings. Moreover, the lack of studies on well-defined groups means that writers have tended to generalise from one population to another, which may not be justified. In this respect, it is worth noting the findings of Peveler et al (101). They assessed patients from the community and found that individuals who attributed their symptoms to food sensitivity suffered less psychological impairment than those who attributed their complaints to stomach or bowel disorders and stress. The estimated rate of psychiatric illness was also relatively low, and they noted a strong relationship between the patients' attribution of symptoms to foods and the clinical judgement of food sensitivity. These findings show that the research relating to allergy clinic patients and people with alleged allergies or various environmental syndromes cannot be generalized to people with food intolerance from the general population. They also suggest that many patients' beliefs about food sensitivity are probably accurate and that claims of widespread misattribution were premature. 


Perhaps the most important and effective remedy for psychologisation in the clinic is for scientists to stop using psychological explanations as a convenient 'dust-bin' for complex cases or troublesome patients. In this hi‑tech age, where doctors depend more and more on the results from laboratory tests and objective clinical signs, a psychological diagnosis can explain the otherwise inexplicable. It is often plausible, difficult to disprove and can be altered to suit changing circumstances. To put it another way, a psychological explanation may be an attractive option for practitioners who lack financial resources and/or time. However, as many writers have pointed out (e.g. 104), it might serve health care professionals better to be 'agnostic' every now and then.

As for the examples in the literature, a more objective and critical approach by reviewers and editors should help to avoid most cases of bias. For instance, McEvedy and Beard's assessment of the outbreak at the Royal Free Hospital was largely based on information from selected case-notes and a highly subjective evaluation of the evidence (9). Their misrepresentation of the data constituted a clear case of psychologisation, reflecting not only their own lack of respect for science but also the editors' failure to address the paper's many flaws.

The same lack of editorial interest in accuracy and balance has allowed more recent articles to exaggerate the role of psychiatric morbidity in CFS (75, 105, 106). Here writers rarely consider alternative explanations for the associated mood disorders, such as the effect of uncertainty, the strain of the illness and the lack of social support (8, 24). This bias illustrates the editors' continued preference for simple, psychological explanations at a time when the evidence points to a heterogeneous population and a multifactorial aetiology (107, 108, 109, 110). In fact, the predilection for simplistic theories and lack of interest in alternative views features in many cases of psychologisation, not just those related to CFS.

People are entitled to their opinion and I am not suggesting that speculation has no place in scientific texts. However, unsubstantiated or poorly supported claims should be acknowledged as such and interpreted with care. In this age of evidence-based medicine, writers must consider all reasonable explanations, and take account of methodological flaws and valid criticisms.

It is my belief that articles with a strong bias towards one explanation and an unbalanced or unfair approach to the available data, do not belong in scientific publications unless marked as an opinion piece. Respect for plausible alternative views and the evaluation of all the relevant evidence are not luxuries, to be indulged in should space permit. A broader editorial policy may produce more readable and provocative articles but it also leads to psychologisation and accordingly, to misunderstanding, inappropriate advice and much needless distress.


1. It is generally accepted that CFS is not a single entity, but that the term covers a number of disorders, all characterised by unexplained fatigue. The evidence supports the view that ME should be regarded as a subgroup of CFS.


1 Goudsmit EM and Gadd R. All in the mind? The psychologisation of illness. The Psychologist 1991; 4: 449-453.

2 Rippere V. The mental state of dismissed patients - an enquiry into dismissal injury. V. Repercussions of psychogenic dismissal. Newsletter of the Society for Environmental Therapies 1992; 12: 138-148.

3 Treasure RAR, Fowler PBS, Millington HT and Wise PH.   Misdiagnosis of diabetic ketoacidosis as hyperventilation syndrome. Br Med J 1987; 294: 630.

4 Brozovic M. With women in mind. Br Med J 1989; 299: 689.

5 Booth G. Psychodynamics in Parkinsonism. Psychosom Med 1948; 10: 1-14.

6  Hay L. You can heal your life. London: Eden Grove, 1988.

7  Millenson JR. Mind matters. Psychological medicine in holistic practice. Seattle: Eastland Press, 1996.

8  Goudsmit EM. The psychological aspects and management of chronic fatigue syndrome. Doctoral Thesis. Brunel University, 1996.

9  Ramsay AM. Myalgic encephalomyelitis and postviral fatigue states, 2nd ed. London: Gower Medical Publishing, 1988.

10 McEvedy CP and Beard AW. Royal Free epidemic of 1955: a reconsideration. Br Med J 1970; 1: 7-11.

11 Crowley N, Nelson M and Stovin S. Epidemiological aspects of an outbreak of encephalomyelitis at the Royal Free Hospital in the Summer of 1955. J Hygiene (Cambridge) 1957; 55: 102-122.

12 Medical Staff of the Royal Free Hospital. An outbreak of encephalomyelitis in the Royal Free Hospital Group, London in 1955. Br Med J 1957; 2: 895-904.

13 Editorial. Br Med J 1978; 1: 1436-1437.

14 Jenkins R. Introduction. In: Jenkins R and Mowbray J (eds):  Post-viral fatigue syndrome. Chichester: John Wiley & Sons, 1991.

15 Leitch AG. The chronic fatigue syndrome reviewed. Proc Roy Coll Physicians Edinb 1994; 24: 480-508.

16 Wessely S. Mass hysteria: two syndromes? Psychol Med 1987; 17: 109-120.

17 Gosling PH. Epidemic malaise. Br Med J 1970; 1: 499-500.

18 Goudsmit EM.  All in her mind! Stereotypic views and the psychologisation of women's illness. In: Wilkinson S and Kitzinger C, eds: Women and health. Feminist perspectives. London: Taylor and Francis, 1994.

19 Sirois F. Epidemic hysteria. In: Roy A, ed: Hysteria. Chichester: Wiley and Sons, 1983.

20 Ramsay AM. Encephalomyelitis in North London. An endemic infection simulating poliomyelitis and hysteria. Lancet 1957; 2: 1196-1200.

21 Scott BD. Epidemic malaise. Br Med J 1970; 1: 170.

22 Innes SGB. Encephalomyelitis resembling benign myalgic encephalomyelitis. Lancet 1970; i: 969-971.

23 Vollmer-Conna U, Lloyd A, Hickie I and Wakefield D. Chronic fatigue syndrome: an immunological perspective. Aust N Z J Psychiatry 1998; 32: 523-527.

24 Wessely S, Hotopf M and Sharpe M, eds: Chronic fatigue and its syndromes. Oxford University Press, 1998.

25 Lennane KJ and Lennane RJ. Alleged psychogenic disorders in  women - a possible manifestation of sexual prejudice.  N Engl J Med 1973; 288: 288-292.

26 Notman M. The psychiatrist's approach. In: Debrovner CH, ed: Premenstrual tension. A multidisciplinary approach. NY: Human Sciences Press, 1982.

27 Goudsmit EM. Psychological aspects of premenstrual symptoms. In: Brush MG and Goudsmit EM, eds: Functional disorders of the menstrual cycle. Chichester: Wiley & Sons, 1988.

28 Roberts H. The patient patients. London: Pandora Press, 1985.

29 Gillespie L. You don't have to live with cystitis. London: Century, 1988.

30  Zola IK. Problems of communications, diagnosis, and patient care: the interplay of patient, physician and clinic organisation. J Med Educ 1963; 38: 829-838.

31 Howell MC. What medical schools teach about women. N Engl J Med 1974; 291: 304-307.

32 Armitage KJ, Schneiderman LJ and Bass RA. Response of physicians to medical complaints in men and women. JAMA 1979; 241: 2186-2187.

33 Ayanian JZ and Epstein AM. Differences in the use of procedures between women and men hospitalized for coronary heart disease. N Engl J Med 1991; 325: 221-225.

34 Steingart R, Packer M, Hamm P, Coglianese ME, Gersh, B et al. Sex differences in the management of coronary heart disease. N Engl J Med 1991; 325: 226-230.

35. Pope JH, Aufderheide TP, Ruthazer R, Woolard RH, Feldman JA et al. Missed diagnoses of acute cardiac ischemia in the emergency  department. N Engl J Med 2000; 342: 1163-1170.

36 Lopez SR. Patient variable biases in clinical judgement: conceptual overview, and methodological considerations. Psychol Bull 1989; 106: 184-203.

37 Himmelhoch J, Pincus J, Tucker G and Detre T. Sub-acute encephalitis: behavioural and neurological aspects. Br J Psychiatry 1970; 116: 531-538.

38 Shilts R. And the band played on. London: Penguin, 1988.

39 Nicholson GA, Wilby J and Tennant C. Myasthenia gravis: the problem of a "psychiatric" misdiagnosis. Med J Aust 1986; 144, 632-638.

40 Hartnell L. Personal view. Br Med J 1987; 294: 1029.

41 Stewart DC and Sullivan TJ. Illness behavior and the sick role in chronic disease. Soc Sci Med 1982; 16: 1397-1404.

42 Easton C. Jacqueline Du Pré. A biography. London: Hodder & Stoughton, 1989.

43 Robinson I. Multiple sclerosis. London: Routledge, 1988.

44 Taylor AG, Skelton JA and Butcher J. Duration of pain condition and physical pathology as determinants of nurses' assessments of patients in pain. Nursing Research 1984; 33: 4-8.

45 Preece PE, Mansel RE and Hughes LE. Mastalgia: psychoneurosis or organic disease?  Br Med J 1978; 1: 29-30.

46 Dakof GA and Mendelsohn GA. Parkinson's Disease: the psychological aspects of a chronic illness. Psychol Bull 1986; 99, 375-387.

47 Kowal SJ. Emotions as a cause of cancer. Psychoanal Rev 1955; 42: 217-227.

48 Langworthy OR. Relation of personality problems to onset and progress of multiple sclerosis. Arch Neurol Psychiatry 1984; 59: 13 28.

49 Philippopoulos GS, Wittkower ED and Cousineau A. The etiological significance of emotional factors in onset and exacerbations of multiple sclerosis. Psychosom Med 1958; 6: 459-474.

50 Surwit RS, Scovern AW and Feinglos MN. The role of behavior in diabetes care. Diabetes Care 1982; 5: 337-342.

51 Panagis DM. Psychological factors and cancer outcome. In:  Cohen J, Cullen JW and Martin LR (eds): Psychological aspects of cancer. New York: Raven Press, 1982.

52 Rodin GM, Craven J and Littlefield C, eds: Depression in the medically ill. An integrated approach. New York: Brunner Mazel, 1991.

53  Yeomans JDI and Conway SP. Biopsychosocial aspects of chronic fatigue syndrome (myalgic encephalomyelitis). J Infect 1991; 23: 263-269.

54 Millon C, Salvato F, Blaney N, Morgan R, Mantero-Atienza E  et al. A psychological assessment of chronic fatigue syndrome/chronic Epstein-Barr virus patients. Psychology and Health 1989; 3: 131-141.

55 Minden SL, Orav J and Reich P. Depression in multiple sclerosis. Gen Hosp Psychiatry 1987; 9: 426-434.

56 American Psychiatric Association. Quick reference to the diagnostic criteria from DSM-IV. Washington: APA, 1994.

57 Dutton DG. Depression/somatization explanations for the chronic fatigue syndrome: a critical review. In: Hyde BM, Goldstein J and Levine P, eds: The clinical and scientific basis of myalgic encephalomyelitis/chronic fatigue syndrome. Ottawa: The Nightingale Research Foundation, 1992.

58 Ray C. Chronic fatigue syndrome and depression: conceptual and methodological ambiguities. Psychol Med 1991; 21: 1-9.

59 Thase ME. Assessment of depression in patients with chronic fatigue syndrome. Rev Infect Dis 1991; 13  Suppl 1: S114-118.

60 Bukberg J, Penman D and Holland JC. Depression in hospitalized cancer patients. Psychosom Med 1984; 46: 199-212.

61 Cavanaugh SA. Depression in the medically ill. In: Judd FK, Burrows GD and Lipsett DR, eds: Handbook of studies on general hospital psychiatry. Amsterdam: Elsevier, 1991.

62 Clark DC, Cavanaugh S and Gibbons RD. The core symptoms of depression in medical and psychiatric patients. J Nerv Ment Dis 1983; 171: 705-713.

63 Frank RG, Beck NC, Parker JC, Kashani JH, Elliott TR et al. Depression in rheumatoid arthritis. J Rheumatol 1988; 15: 920-925.

64 Krupp LB, Alvarez LA, LaRocca NG and Scheinberg LC.  Fatigue in multiple sclerosis. Arch Neurol 1988; 45: 435-437.

65 Starkstein SE, Preziosi TJ, Forrester AW and Robinson RG. Specificity of affective and autonomic symptoms of depression in Parkinson's disease. J Neurol, Neurosurg Psychiatry 1990; 53: 869-873.

66 Katon W, Buchwald DS, Simon GE, Russo JE and Mease PJ. Psychiatric illness in patients with chronic fatigue and those with rheumatic arthritis. J Gen Intern Med 1991; 6: 277-285.

67 Johnson SK, DeLuca J and Natelson BH. Assessing somatization disorder in the chronic fatigue syndrome. Psychosom Med 1996; 58: 50-57.

68 Krupp L, Sliwinski M, Masur DM, Friedberg F and Coyle PK. Cognitive functioning and depression in patients with chronic fatigue syndrome and multiple sclerosis. Arch Neurol 1994; 51: 705-710.

69 Pepper CM, Krupp LB, Friedberg F, Doscher C and Coyle PK.  A comparison of neuropsychiatric characteristics in chronic fatigue syndrome, multiple sclerosis and major depression. J Neuropsychiatry Clin Neurosciences 1993; 5: 200-205.

70 Walford GA, McC Nelson, W and McCluskey DR. Fatigue, depression, and social adjustment in chronic fatigue syndrome. Arch Dis Child 1993; 68: 384-388.

71 Wessely S and Powell R. Fatigue syndromes: a comparison of chronic "postviral" fatigue with neuromuscular and affective disorders. J Neurol Neurosurg Psychiatry 1989; 52: 940-948.

72  Wood GC, Bentall RP, Gopfert M and Edwards RHT. A comparative psychiatric assessment of patients with chronic fatigue syndrome and muscle disease. Psychol Med 1991; 21, 619-628.

73 Abbey SE. Somatization, illness attribution and the sociocultural psychiatry of the chronic fatigue syndrome. In: Bock GR and Whelan J, eds: Chronic fatigue syndrome.  Chichester: Wiley and Sons, 1993

74 David AS. Postviral fatigue syndrome and psychiatry. Br Med Bull 1991; 47: 966-988.

75 Thomas PK. The chronic fatigue syndrome: what do we know? Br Med J 1993; 306: 1557-1558.

76 Brahams D. Fatal case of undiagnosed diabetes. Lancet 1990; 335 i: 652.

77 Longden D. Diana's story. London: Bantam Press, 1989.

78 Lerner MJ and Simmons CH. Observer's reaction to the  "innocent victim": compassion or rejection? J Pers Soc Psychol 1966; 2: 203-210.

79 Taylor SE. Adjustment to threatening events. A theory of cognitive adaptation. Am Psychol 1983; 38, 1161-1173.

80 Fox BH. Psychogenic factors in cancer, especially its incidence. In: Maes S, Spielberger CD, Defares PB and Sarason, ID, eds: Topics in health psychology. Chichester: Wiley and Sons, 1988.

81 Ramirez AJ, Craig TKJ, Watson JP, Fentiman IS, North WRS and Rubens RD.  Stress and relapse of breast cancer. Br Med J 1989; 298: 291-293.

82 Turnquist DC, Harvey JH and Andersen BL. Attributions and adjustment to life-threatening illness. Br J Clin Psychol 1988; 27: 55-65.

83 Sontag S. Illness as metaphor. London: Penguin, 1977.

84 Wilkinson S and Kitzinger C. Towards a feminist approach to breast cancer. In: Wilkinson S and Kitzinger C, eds: Women and Health. Feminist Perspectives. London: Taylor and Francis, 1994.

85  Wortman CB and Dunkel-Schetter C. Interpersonal relationships and cancer: a theoretical analysis. J Soc Issues 1979; 35, 120 155.

86 Howard LM and Wessely S. Psychiatry in the allergy clinic: the nature and management of patients with non-allergic symptoms. Clin Exp Allergy 1995; 25: 503-514.

87 Black DW, Rath A and Goldstein RB. Environmental illness. A controlled study of 26 subjects with '20th century disease'. JAMA 1990; 264: 3166-3170.

88 Simon GE, Daniell W, Stockbridge H, Claypoole K and Rosenstock L. Immunologic, psychological, and neuropsychological factors in multiple chemical sensitivity. A controlled study. Ann Int Med 1993; 119: 97-103.

89 Brodsky  CM. 'Allergic to everything': a medical subculture. Psychosomatics 1983; 24: 731-744.

90 Rix KJB, Pearson DJ and Bentley SJ. A psychiatric study of patients with supposed food allergy. Br J Psychiatry 1984; 145: 121-126.

91 Stewart DE and Raskin J. Psychiatric assessment of patients with "20th century disease" ("total allergy syndrome"). Can Med Assoc J 1985; 133: 1001-1006.

92 Brostoff J and Gamlin L. The Complete Guide to Food Allergy and Intolerance. London: Bloomsbury, 1989.

93 Anthony H, Birtwistle S, Eaton K and Maberly J. Environmental medicine in clinical practice. Southampton: BSAENM Publications, 1997.

94 Radcliffe MJ, Ashurst P and Brostoff J. Unexplained illness: the mind versus the environment. J R Soc Med 1995;  88: 678-679.

95 Christman NJ, McConnell EA, Pfeiffer C, Webster KK, Schmitt M and Ries, J. Uncertainty, coping and stress following myocardial infarction: transition from hospital to home.  Research in Nursing and Health 1988; 11: 71-82.

96 Wineman NM. Adaptation to multiple sclerosis: the role of social support, functional disability, and perceived uncertainty. Nursing Research 1990; 39: 294-299.

97 Rippere V. The mental state of dismissed patients - an enquiry into dismissal injury. II. Dismissal as unfinished business. Newsletter of the Society for Environmental Therapies 1991; 11: 72-79.

98 Woodward RV, Broom DH and Legge DG. Diagnosis in chronic illness: disabling or enabling - the case of chronic of chronic fatigue syndrome. J R Soc Med 1995; 88: 325-329.

99 Simon GE, Katon WJ and Sparks PJ. Allergic to life: psychological factors in environmental illness. Am J  Psychiatry 1990; 147: 901-906.

100 Hunter JO. Food allergy - or enterometabolic disorder? Lancet 1991; 338: 495-496.

101 Peveler R, Mayou R, Young E and Stoneham M. Psychiatric aspects of food-related physical symptoms: a community study. J Psychosom Res 1996; 41: 149-159.

102 Howard L and Wessely S. The psychology of multiple allergy. Br Med J 1993; 307: 747-748.

103. Jacobsen MB, Aukrust P, Muller F, Ueland T, Bratlie J et al.  Relation between food provocation and systemic immune activation in patients with food intolerance. Lancet 2000; 356: 400-401.

104 Guggenbuhl-Craig A and Micklem N.  No answer to Job: reflections on the limitations of meaning in illness. In:  Kidel M and Rowe-Leete S, eds: The meaning of illness. London: Routledge, 1988.

105 Barsky AJ and Borus JF. Functional somatic syndromes. Ann Int Med 1999; 130: 910-921.

106 Wessely S, Nimnuan C and Sharpe M.  Functional somatic syndromes: one or many? Lancet 1999; 354: 936-939.

107 Christodoulou C, DeLuca J, Johnson SK, Lange G and Natelson BH. Efforts to reduce heterogeneity in chronic fatigue syndrome. In: Yehuda S and Mostofsky DI, eds: Chronic fatigue syndrome. New York: Plenum Press, 1997.

108 Friedberg F. A subgroup analysis of cognitive-behavioral treatment studies. J Chronic Fatigue Syndrome 1999; 5: 3-4, 149-159.

109 Ablashi DV, Eastman HB, Owen CB, Roman MM, Friedman J et al. Frequent HHV-6 reactivation in multiple sclerosis (MS) and chronic fatigue syndrome (CFS) patients.  J Clin Virol 2000; 16: 179-191.

110 Paul L, Wood L, Behan WMH and Maclaren WM. Demonstration of delayed recovery from fatiguing exercise in chronic fatigue syndrome. Eur J Neurol 1999; 6: 63-69.


I wish to thank all those who helped me with the preparation of the manuscript, especially Ms. Mary Sullivan MSc, Mrs Sandra Howes and Mrs. Doris Jones MSc.


Copyright EM. Goudsmit, PhD C. Psychol. August 2000 ©
Editor ME and CFS Capita Selecta Quarterly

All rights reserved. This article may not be reproduced without
permission from the author. See the
full copyright notice.


Be sure to see the many other valuable articles at our Main M.E. Home Page

button - top