The report of the inquiry into the tragic death of Dakhsa Emson, a doctor who killed herself and her child (North East London Strategic Health Authority, 2003), has highlighted the need for psychiatrists to take the mental health of doctors seriously. Treating doctors for mental disorders can be problematic in terms of conflicting loyalties and our own emotional reactions to illness in professionals; there may also be real institutional and managerial complications, such as when the patient is a colleague. These complications also raise ethical issues, which I review here. I conclude with some thoughts about treatment, prevention and education.Footnote 1
Medicine is potentially bad for your health. Doctors suffer increased rates of somatic and social dysfunction (Reference Baldwin, Dodd and WrateBaldwin et al, 1997a ) and higher levels of fatigue than the general population (Reference Hardy, Shapiro and BorrillHardy et al, 1997). The greatest impact of being a professional carer seems to be on mental health, where the practice of medicine is associated with increased rates of minor and major psychiatric illness (Box 1) (Reference Sonneck and WagnerSonneck & Wagner, 1996; Reference Graham and RamirezGraham & Ramirez, 1997; Reference Wall, Bolden and BorrillWall et al, 1997). Depression and substance misuse, in particular, are thought to be the cause of the established increase in rates of suicide in doctors (Reference Hawton, Clements and SakarovitchHawton et al, 2001).
• Physical illness (e.g. cancer, epilepsy) and its psychological effects on the doctor's identity and practice
• Psychiatric and psychological conditions that may affect an individual's fitness to practise: depression, bipolar illness and substance misuse
• Stress and burnout: work, family, life, and compassion failure
• Specific problems for psychiatrists: stigma, hypocrisy, anxiety, lack of evidence base in psychiatry, and increased risk of psychiatric problems
The increased prevalence of physical and psychological morbidity in doctors has been studied extensively, mainly in relation to work stress. Rates of stress are high in all doctors (both hospital doctors and general practitioners), but junior doctors and female doctors seem particularly at risk (Reference RoutRout, 1999; Reference Bogg, Gibbs and BundredBogg et al, 2001). The predictors of work stress are multiple and complex (Reference Linzer, Gerity and DouglasLinzer et al, 2002) and they affect job satisfaction (Reference Swanson, Power and SimpsonSwanson et al, 1996). There also seems to be a relationship between stress and ‘burnout’, a syndrome characterised by emotional exhaustion, depersonalisation and low personal accomplishment (Reference McManus, Winder and GordonMcManus et al, 2002).
Working in mental health seems (both predictably and ironically) to be especially hazardous for the mental health of psychiatric staff (for a useful review see Reference Walsh and WalshWalsh & Walsh, 2001). Reference Guthrie, Tattan and WilliamsGuthrie et al (1999) found high rates of stress and burnout in psychiatrists, especially juniors. They note that dealing with violent patients is highly stressful, regardless of grade, which is consistent with the literature on the relationship between fear exposure and traumatic stress reactions. This literature might be particularly significant for those branches of psychiatry in which the risk of violent assaults by patients is increased, for example intensive care, emergency clinics, community teams who do domiciliary visits and residential forensic psychiatry.
Managing mental illness in doctors is also problematic because doctors are ‘poor’ patients. Maladaptive health behaviours start early in the career; Reference Baldwin, Dodd and WrateBaldwin et al (1997b) found that most junior doctors self-medicated, rarely sought a formal consultation for health problems and continued to go to work when unfit. One-third had no general practitioner; most had no idea of the role of occupational health services. The practice of self-medication may be a risk factor for later substance misuse, which is especially common in anaesthesiology, emergency medicine and psychiatry (Reference Bennet and O'DonovanBennet & O'Donovan, 2001).
Implications for treatment
The subjective experience of being ill is not taught or much discussed in medical school, and most of us find out the hard way through being ill ourselves (as attested to by accounts in the ‘Personal view’ column in the BMJ). For many doctors, illness is something that not only has to be managed and overcome as quickly as possible, but also essentially happens to other people. Wishful thinking and emotional distancing from distress are common methods for doctors to cope with work stress, but these strategies are actually associated with increased stress and risk of mental health problems, presumably because they only work in the short term (Reference Tattersall, Bennett and PughTattersall et al, 1999; Reference Tyssen and VaglumTyssen & Vaglum, 2002). As a result of such attitudes doctors may be reluctant to admit that they are ill and may be especially unwilling to acknowledge that their ill health is affecting their performance at work. In one study comparing doctors with pilots, doctors were less likely to report fatigue and less able to discuss making mistakes at work (Reference Sexton, Thomas and HelmreichSexton et al, 2001).
These findings indicate that doctors may experience high levels of shame associated with being ill, perhaps more than the general population. This is likely to be worse when the problems are psychological in nature. Doctors, like the rest of the population, are affected by stigmatising images of mental illness and its treatment (Reference Mukherjee, Fialho and WijetungeMukherjee et al, 2002). Psychiatrists are likely to be especially reluctant to admit that they are suffering from the same problems as their patients, partly because they too are affected by stigma and partly because they know how poor services are in many areas. How many psychiatrists would care to be admitted to the wards they look after? Psychiatrists may also feel an increased sense of shame that they have not been in total control of their feelings or thoughts, that they have not been able to maintain mastery over their own minds.
There are other reasons for thinking that shame is a key issue affecting the treatment of doctors with mental health problems. Personality traits associated with vulnerability to depression are common in medical students, especially self-criticism and dependency (Reference Firth CozensFirth Cozens, 1992; Reference Brewin and Firth CozensBrewin & Firth Cozens, 1997). Prospective studies indicate that self-criticism is associated with depression, particularly in male doctors, which suggests that doctors tend to place the locus of control firmly in the self, so that they see themselves as personally responsible for all events, both negative and positive. The creation in the mind of such an omnipotent self is a defence against feeling overwhelmed by distress or need of care; the choice of medicine as a profession is then a mental ‘insurance policy against catastrophe’ (Reference ClarkClark, 1995).
The ethical duties to a ‘medical’ patient (a doctor) are the same as those to a ‘non-medical’ patient (a member of the general public), but the therapeutic alliance that gives rise to those duties may be more difficult to achieve, resulting in particular ethical tensions (Box 2).
• A psychiatrist asked to assess a sick doctor must act impartially and be unbiased by personal responses to a doctor's problems
• The decision of an assessing psychiatrist may remove a fellow doctor's livelihood
• Assessing and treating psychiatrists cannot guarantee a sick doctor absolute confidentiality
• A treating psychiatrist has a duty to report a poorly performing doctor, with or without that doctor's permission
• Should doctor-patients, particularly psychiatrists, be treated out-of-area or privately, to protect their identity?
Disclosure and assessment
The first ethical dilemma faced by psychiatrists relates to the identification of problems and the need for action in situations where a fellow doctor (especially a psychiatrist) appears not to be aware of their own problems, or is reluctant to accept the reality or severity of their mental difficulties. As doctors, psychiatrists are under an obligation to report poorly performing doctors (General Medical Council (GMC), 1992), and they may therefore feel conflicting loyalties. Doctor-patients know this, and may be reluctant to admit the extent or severity of their difficulties.
Dilemmas for medical colleagues (and patients) usually centre on the need to disclose their concerns, either in the face of the doctor-patient's refusal or without their knowledge, both of which may result in conflict, anger, distress and deception. The variety of formal procedures that are now available (such as ‘whistle-blowing’, ‘poorly performing doctor’ procedures, appraisal and revalidation) may make resolution of such situations easier, but arranging an assessment may be especially problematic with very senior colleagues and with trainees. Sometimes the multiplicity of procedures involving trusts, deaneries and the GMC can cause confusion and increased stress for doctor-patients.
Psychiatrists may be asked to assess doctors with mental health problems in a variety of settings and for a variety of purposes. The most common situation will be provision of an assess ment at the request of a third party such as a trust's occupational health service or the GMC. Here the doctor being assessed cannot assume that the psychiatrist will provide a report that is helpful to him, nor that the interview is entirely confidential. In this sense, these assessments are like any other reports prepared at the request of third parties, and psychiatrists will need to give the usual warnings about the limits of confidentiality and a duty to disclose risk. Absolute confidentiality cannot be guaranteed at the assessment stage, although it may be possible for the psychiatrist and the doctor-patient to negotiate how and what information will be disclosed, and to whom.
Assessing psychiatrists need to be thoughtful about their own personal responses to the assessment process. It is easy to both over- and under-identify with the doctor being assessed, especially with a fellow psychiatrist. The duty to act impartially in the pursuit of justice and beneficence means that taking one's own feelings seriously is part of the assessment process in order to avoid being either too punitive in response or too dismissing of difficulties. Communication with other evaluators and discussion with third parties who know the doctor in different ways may be helpful.
Another ethical issue that arises during assessment relates to the pursuit of justice and equitable treatment. There is an absence of an evidence base addressing the relationship between mental disorder and work performance, which means that judging fitness to practise can have a potentially arbitrary quality. For example, in some cases of mental illness, work will be an important support structure during an episode, and staying in work (in some form) may help an individual to recover. In other cases, it will be clear that the mental illness or psychological problem affects fitness to practise in an obvious way, and continued employment will not be possible. However, there are some types of psychological and psychiatric problems whose impact on fitness to practise is unclear; and given that lack of clarity, it would seem unjust to remove a person's livelihood (with the accompanying impact on their mental health) without good reason. For example, a psychiatrist may be asked to comment on whether paedophilia affects fitness to practise as an orthopaedic surgeon; or whether a trainee with a history of bipolar affective disorder should pursue a career as a psychotherapist. These are decisions with enormous consequences for the doctor-patients concerned, and psychiatrists’ opinions often carry great weight.
Confidentiality and competence
Many of the ethical issues that arise in the treatment of sick doctors resemble those that arise during assessment. In both cases, issues of context and role identity can lead to dilemmas about consent, disclosure of information and dual loyalties.
A doctor who is referred for psychiatric treatment via primary care may refuse to let the treating psychiatrist speak to his employers about his fitness to practise, or to family members to find out more details about symptoms. He may or may not be competent to refuse this aspect of treatment; if he is competent to refuse, then the psychiatrist will have to weigh up the risks entailed in not discussing the case with third parties. The duties of a doctor registered with the GMC would support breaches of confidentiality in these circumstances, both on the grounds of the prevention of harm and on the duty to disclose information about poorly performing colleagues (General Medical Council, 1992). Of course, doctor-patients should know this better than anyone, but it is remarkable how many doctors (both as patients and treaters) still hope that the duty of confidentiality is absolute.
The more common issue around disclosure relates to providing progress reports for third parties. If a doctor is in treatment for a problem that has impaired fitness to practise, then it is likely that employers or the GMC will seek feedback on progress (or lack of it). In such circumstances, if the patient gives consent to disclosure then there is no dilemma. Depending on the situation, the treating psychiatrist may:
a speak directly to the third party, or
b pass on their views through another professional such as the general practitioner or supervisor (in GMC cases).
In practice, it is usually best (and fairest) if the evaluation and treatment roles are separated. However, this may not solve the ethical tension; if the doctor-patient refuses to give consent to any disclosure, then the treating psychiatrist will be faced with the same dilemma described above, namely whether any risk exists that justifies disclosure.
Protection of ‘public privacy’
Such dilemmas assume that the patient is competent to consent or refuse. However, patients may become so ill that they require treatment in hospital. The treating psychiatrist's duty to benefit the patient may entail use of the Mental Health Act 1983 if the patient is sufficiently ill. In this situation, treating psychiatrists may feel reluctant to detain doctor-patients in their own local units, where they may meet patients they have treated. The ethical issue here is one of fairness: doctors get access to protection of their privacy that other mental health service users do not necessarily get. Some trusts do have policies that make provision for out-of-area treatment of doctors so that they can be treated away from their own professional setting. However, there is an ethical question about the extent to which protecting doctors in this way reinforces stigma and shame, even while it protects the doctor's identity as a patient. Do doctors have a duty to use the services they recommend for their patients?
There is another interesting ethical question about resources for treatment. In an effort to prevent doctor-patients feeling embarrassment or shame in receiving treatment from local facilities, they are often under pressure to seek help in the private sector. However, private medical insurance often will not pay for extended psychiatric treatment, especially for substance misuse problems (Reference Wall and AppelbaumWall & Appelbaum, 1998); nor can it be assumed that the quality of treatment in the private sector is the same as in the National Health Service. Financing such treatment can be a particular problem for junior doctors and for those who are barred from employment because of their mental condition. There is good evidence that self-help groups are a major factor in recovery from addiction problems in doctors (Reference BrookeBrooke, 1997), but gaining access to these groups can be difficult since they are almost exclusively provided by private treatment centres.
Risk and fitness to work
The chief professional concern is whether the sick doctor poses a risk to others as a result of the mental disorder. However, the assessment of risk in relation to mental disorder and decision-making based on that assessment are fraught with ethical dilemmas: again, these are mainly to do with justice and equitable treatment and the lack of evidence to justify decisions that have huge consequences. For example, it seems plausible to argue that doctors addicted to alcohol will be a risk at work and unable to practise. However, anecdotally at least, it appears that it is all too possible to carry on being an adequate practitioner while being an addict. I make this point not to suggest that addiction is not harmful or should not be treated, but only to demonstrate that it may not be all that easy to quantify the risk to others posed by a mentally disordered doctor. If a doctor drops out of psychiatric treatment, is the unknown risk sufficient to justify informing someone of this? And if so, who? There are many ethical concerns about risk assessment in ordinary psychiatric populations (e.g. whether it is sufficiently accurate to justify breaches of confidentiality or detention; Reference Logan, Adshead and BrownLogan, 2003) and these will apply in the same way when assessing risk in patients who are doctors.
The purpose of treatment
Another issue that is raised by treatment relates to its purpose. Is it to make the doctor safer or better? Consider again the alcohol-misusing doctor who can be helped to abstinence with treatment. Employers and the GMC may be happy to employ someone who is abstinent, and the doctor herself may be keen to resume work. However, the psychiatrist may not be so convinced that all is well if it is obvious that underlying emotional problems remain that are powerful risk factors for relapse. Attributing mental disorder or addiction problems simply to external ‘stress’ can lead to the simplistic assumption that now the ‘stressor’ is altered, all is well. Work stress may cause mental disorder (Box 3), but in some cases what the stress reveals is an underlying vulnerability, which will not be put right by simply altering external circumstances. The treating psychiatrist may face professional and managerial pressures that seek to minimise the doctor's problems, perhaps to solve a financial or workplace difficulty or to defend against the reality of the doctor's distress.
• Treatment and care of patients
• Lack of time for patient care
• Professional isolation
• Complex patients
• Intensive contact with very ill patients
• Team conflicts
• Lack of autonomy
• Increased criticism and expectations from the public
• Role conflict between work and family
• Long hours
• Imbalance with between work and family time
Finally, and relatedly, there is the vexed question of the use of terminology. Many doctors will be happier to be diagnosed as having burnout (Box 4) than clinical depression or anxiety. Burnout appears to be a less stigmatising label and to be viewed as a largely occupational disorder that is comparatively minor and easily put right. However, low personal accomplishment and emotional exhaustion are also symptoms of depression, and depersonalisation is a dissociative symptom associated with anxiety disorders. Treating psychiatrists may come under pressure from doctor-patients and their colleagues to use terminology that minimises the doctor's problems. Treating psychiatrists also have to be thoughtful about their own countertransference to sick doctors when they make diagnoses and recommend treatment; they too may be susceptible to minimising doctors’ difficulties.
• Low personal accomplishment
• Emotional exhaustion
Implications for professional development
The mental health of many doctors could be improved with some simple occupational interventions: more support for junior staff and time for discussion, less tolerance of sleep deprivation, and strategies to balance work and family life (Reference Paice, Rutter and WetherallPaice et al, 2002a ). There is reason to think that less strictly hierarchical management structures might make it easier for doctors to discuss mistakes and might promote learning in health systems, rather than scapegoating and focusing on individual errors (Reference Sexton, Thomas and HelmreichSexton et al, 2001). This is an important area in healthcare management, because medical errors account for a significant number of deaths each year, and work stress is related proportionately to error rate (Reference Houston and AlltHouston & Allt, 1997; Reference Firth CozensFirth Cozens 2003).
Attitudes to caring
Many of the practical and ethical dilemmas that arise when treating sick doctors do so because of the values, beliefs and attitudes of the medical profession. We know that some people become professional carers as a response to their own experience of being cared for in early childhood or difficult relationships with their carers (Reference Firth CozensFirth Cozens, 1992). Insecure attachment in childhood, especially where the child grows up in a frightening environment, increases the chance of that child developing a ‘compulsive caregiving’ style in adulthood (Reference BowlbyBowlby, 1969). Becoming a professional carer may be a psychological way of managing unconscious anxieties about being needy and dependent. Such doctors may then invest enormous amounts of psychological time in ‘not being patients’. If a person's principal psychological aim in being a doctor is not to be a patient (i.e. needy and vulnerable), then when the ordinary stressful events of life occur that must naturally generate those feelings, the doctor will be under both internal and external pressure to escape the feelings or abolish them. A vicious psychological circle is set up whereby the doctor's professional and personal identity and sense of value are under attack, both from his anxiety about admitting distress and from the professional and managerial culture that seeks to minimise distress.
Attitudes to distress
Three other issues are germane here: cultural associations of distress with weakness and vulnerability; contempt for, and dismissal of, feelings of distress and need; and associated cultural behaviours that encourage the external ablation of internal distress with alcohol or other substances. The first two are also associated with traditional stereotypes of masculinity and may account for the continuing suspiciousness towards women in medicine (Reference Cooke and HutchinsonCooke & Hutchinson, 2001) and the high rates of mental distress and suicide in female doctors (Reference Lindeman, Läärä and HaakoLindeman et al, 1996). Although all doctors need to develop some degree of detachment and capacity to self-soothe as a means of coping with distressing events, this is not the same as the dismissal or disavowal of feelings. A fairly recent study of general practitioners suggests that doctors still feel the need to portray an image of perfect health to patients and colleagues (Reference Thompson, Cupples and SibbettThompson et al, 2001). Since few of us will have perfect health in a long and active life, this suggests that many doctors will come under pressure to portray themselves as well when they are not. If doctors do not take their own distress seriously, they are unlikely to be able to take the distress of their patients seriously.
To avoid sickness in doctors, we need to take their task of care-giving seriously, as seriously as curing people. It would be useful to know whether rates of psychiatric morbidity are higher in doctors working with patients who may never get better. These doctors may have to do more caring than curing, and this may be especially stressful. Some of the literature about parenting skills may be useful here: we know that the capacity to care effectively for dependent children is influenced by one's own recalled experience of being cared for in childhood (Reference van Ijzendoornvan Ijzendoorn, 1995). Why should this not be true for doctors? Their capacity to care for others will, however, be influenced not only by their childhood experience, but also by their experience with the consultants who train them. If their trainers have a dismissive or contemptuous attitude to caring, then it will be hard for junior doctors not to adopt the same approach, not only to their patients, but to themselves when they are in distress (Reference Paice, Moss and HeardPaice et al, 2002b ).
Medicine and medical politics need to change their attitudes to illness and dependency. Central allocation of time resources means that there is less and less space and time for caring and communicating, which is what users of health services say they want most from their doctors. Training in communication skills (especially about difficult or painful issues) is still not obligatory in medicine, even in psychiatry and psychotherapy! Doctors famously won't go to staff groups or staff support (Reference MainMain, 1957), as though therapeutic authority rests on not having any problems, rather than modelling dealing with problems and loss successfully. It is almost as if doctors would rather become alcoholic and misuse drugs than admit need; and sometimes they can only act out their distress through rule breaking. A number of established networks offer support to healthcare professionals (Box 5), and we would do well to seek help from them.
• The Sick Doctor's Trust Tel.: 0870 444 5163
• BMA Counselling Service Tel.: 0845 920 0169
• Doctors’ Support Line Tel.: 0870 765 0001
• Doctors’ Support Network Tel.: 0870 321 0642
• MedNet for the London Deanery
• National Clinical Assessment Authority
Although there are managerial and policy developments that could improve mental health for many doctors, the key message is that doctors are no more immune from becoming mentally ill than any other member of the public. The report on the treatment of Daksha Emson suggested that we need better resources for assessing and treating sick doctors and that occupational health services in particular need to be strengthened, especially in terms of psychiatric care.
A recent suggestion is that we should move away from a disease model and focus on positive functioning as a means of promoting well-being in doctors (Reference Yamey and WilkesYamey & Wilkes, 2001). Although this sounds kind and respectful, such a suggestion ignores the fact that doctors do become mentally ill and will need all the same resources and help that other patients receive. The real challenge is to be able to develop, present and maintain a therapeutic identity that doesn't require doctors to be perfectly pure, perfectly good or perfectly healthy.