10.1192/apt.3.5.275Access the most recent version at DOI: 1997, 3:275-281.APT 

E. Guthrie and D. BlackPsychiatric Disorder, Stress and Burn-out

Referenceshttp://apt.rcpsych.org/content/3/5/275.citation#BIBLThis article cites 0 articles, 0 of which you can access for free at:

permissionsReprints/

permissions@rcpsych.ac.ukto To obtain reprints or permission to reproduce material from this paper, please write

to this article atYou can respond http://apt.rcpsych.org/cgi/eletter-submit/3/5/275

from Downloaded

The Royal College of PsychiatristsPublished by on February 17, 2013http://apt.rcpsych.org/

http://apt.rcpsych.org/site/subscriptions/go to: Adv. Psychiatr. Treat. To subscribe to

Advances in Psychiatric Treatment (1997), vol. 3, pp. 275-281

Psychiatric disorder, stress andburn-out

E. Guthrie & D. Black

This paper will focus primarily on the most recentresearch studies involving doctors in the UK. Wewill not discuss the treatment of doctors who arepsychiatrically unwell as this is covered in otherpapers elsewhere in this issue. Before reviewingspecific studies, we will consider the measurementof psychiatric disorder, stress and burn-out.

Measurement and definitions

Psychiatric disorder

Studies of psychiatric disorder in doctors havemainly been confined to the study of minorpsychiatric morbidity. Other indicators of moresevere disorder, such as suicide, have been used,but not as frequently. Most UK studies have usedGoldberg's General Health Questionnaire to screen

for psychological morbidity. Different forms of thequestionnaire and different cut-off points havebeen used, which can make comparisons difficult.The GHQ-12 (Goldberg & Williams, 1988) has beenmost widely employed with a conservative cutoff of three-quarters. Borrill et al (1996) haverecently validated the GHQ-12 in a large sampleof health service employees using the ClinicalInterview Schedule. They found that three-quarters was the best cut-off score yielding anestimated sensitivity of 73% and specificity of 86%.These validity coefficients match those reportedin other studies on the GHQ and are acceptable.

There are very few studies in doctors in which astandardised semi-structured interview has beenused to assess psychiatric disorder. One such studyis that by Wrate & Baldwin in Edinburgh (see pp.

290-296, this issue) who are following-up a groupof approximately 200 junior doctors, who wereassessed in detail as medical students. The studyis still in progress. When completed, it will be themost detailed study of junior doctors that has everbeen undertaken.

Burn-out

Burn-out is a constellation of symptoms whichdevelop in professionals in relation to work.Previously hard-working, diligent and committedindividuals become increasingly dissatisfied,disillusioned and cynical in relation to work andother aspects of their lives. Doctors who previouslycared about their patients find themselvesdreading clinical work and becoming more andmore detached and uncaring towards patients.Pines et al (1981), highlighted the importantdepressive component of the condition andreferred to:

physical depletion, feelings of helplessness andhopelessness, emotional drain, the development ofnegative attitudes to work, life and other people ... asense of distress and failure in the quest for ideals.

Freudenburger & Richelson (1980) highlightedthe sense of the loss of idealism in a previouslyenergised and committed individual. Theydescribed it as:

a state of fatigue or frustration brought about bydevotion to a cause, way of life, or relationship thatfailed to produce the expected reward.

Edelwich & Brodsky (1980) defined burn-out as:

a progressive loss of idealism, energy, purpose, andconcern as a result of conditions of work.

E. Guthrie is a senior lecturer in liaison psychiatry at the University of Manchester, School of Psychiatry and Behavioural Sciences,Rnwnsley Building, Manchester Royal Infirmary, Oxford Road, Manchester M13 9WL, interested in psychotherapy and the nature ofstress and psychological morbidity in medical students. D. Black is a consultant psychiatrist at Hope Hospital, Salford. Dr Black hasbeen involved in a five-year project examining the prevalence of psychological morbidity in medical students. She is also interestedin research into the effects of day hospital and community treatment for patients with psychiatric disorder.

APT (1997), vol. 3, p. 276 Guthrie & Black

It can be argued that doctors are particularlysusceptible to burn-out because of the high idealsof the profession and the long working hours. Thehealth service reforms have also placed additionalpressures upon doctors, increasing their workload,decreasing a sense of support, increasing patientexpectation, and greater support for patientswishing to make complaints against doctors.

The two most widely used measures of burnout are the Maslach Burn-Out Inventory (MBI;Maslach & Jackson, 1986) and the Burn-OutMeasure (Pines et al, 1981). The Burn-Out Measurehas not been used on a British population ofdoctors. It was, however, used recently in the USAin a sample of 440 middle-aged physicians whohad previously been screened with a personalitytrait measure (Minnesota Multiphasic PersonalityInventory (MMPI); Graham, 1987) 25 years earlier(McCranie & Brandsma, 1988). Not surprisingly,given the large time difference, little associationwas found between the scores on the MMPI whenthe physicians were young men and the burn-outratings when they were middle-aged professionals.

The MBI has been used in several studies ofhealth professionals, including doctors, althoughit has been used in only one large-scale study ofUK doctors (Ramirez et al, 1996). The MBI has threesubscales: emotional exhaustion, depersonalisation(feelings of detachment and callousness towardspeople at work) and personal accomplishment(feelings of competency and achievement at work).The total score for each subscale is categorised'low', 'average' or 'high' according to predetermined

cut-off scores based on normative data from asample of American health professionals. A highdegree of burn-out is indicated by high scores onthe emotional exhaustion and depersonalisationsubscales and low scores on the personal accomplishment subscale.

There is considerable overlap between theemotional exhaustion subscale of the MBI and thesymptomatology of depression. Glass et al (1996)conducted multiple regression analyses on each ofthe MBI subscales using the Beck DepressionInventory as one of the predictor variables in asample of 162 nurses. They found that over 19% ofthe variance in the emotional energy subscale wasattributable to the scores on the Beck DepressionInventory, whereas the percentages of thevariance on the depersonalisation and personalaccomplishment subscales were only 3.9 and 6.5%,respectively.

Thus, burn-out is not just an alternative way ofviewing depression. Although the emotionalexhaustion subscale and depressive symptomalogyshare appreciable variance, the other two subscalesappear to measure different entities.

Stress

Stress is an ill-defined term and its use has beenheavily criticised. The word 'stress' is derived fromthe Latin word stringere which means 'to drawtight'. Most theories of stress involve two factors:

an interaction between the individual and theenvironment, which results in a demand, and afailure of the individual to tolerate or integrate thedemand. Most studies of stress in doctors havefocused upon the first part of the concept; studyingwhich particular aspects of a job create the moststress (British Medical Association, 1992). Thereare few well-recognised measures of stress andmany studies have developed their own ratingscales in which respondents are asked to ratevarious aspects of work in terms of how stressfulthe respondents perceive them to be. It is difficultto make comparisons between studies whendifferent measures have been used.

A recognised measure of stress that has beenused in UK studies of doctors is the OccupationalStress Indicator (Cooper et al, 1988). This is dividedinto six component scales, each of which is dividedinto a set of subscales. The scales are: sources ofpressure in the job, coping skills, type A behaviourpattern, job satisfaction, locus of control andmental and physical ill-health.

There are some concerns regarding the psychological subscales of this questionnaire. Since thismeasure was developed, interest in issues to dowith locus of control have waned. The mentalhealth subscale is not a well-recognised measureof psychological distress, but it has been validatedagainst the Crown-Crisp Experiential Index (1979).

Firth (1986) has also developed a checklist formeasuring aspects of stress, particularly in relationto doctors. This measure was used in her study ofjunior hospital doctors and respondents are askedto rate on a four-point scale various aspects ofmedical-related work.

Morbidity in doctors

Although doctors have lower rates of many kindsof physical illness than the general population, paststudies of sickness in the medical profession havehighlighted three particular areas of concern.Doctors have higher rates of suicide, psychiatricillness and possibly alcohol and drug misuse thanthe general population.

The rate of suicide in doctors is three times thatin the general population, but most of the increasedrate is accounted for by women doctors (British

Psychiatrie disorder, stress & burn-out APT (1997), vol. 3, p. 277

Medical Association, 1993).In 1968women doctorswere almost four times more likely to killthemselves than the general population. Figuresreported by Kelly et al (1995)for the period 1982-1992 are very similar, with women doctors being3.22 times more likely to kill themselves than thegeneral population. Male doctors are 1.4 timesmore likely to take their own lives (Box 1).

There is some evidence that the suicide rates fordoctors peak in mid-life rather than showing alinear increase with age as for the generalpopulation (Steppacher & Mauser, 1974).There isalso evidence that suicide in doctors is under-reported, as many deaths are not classified assuicide but as 'accidental poisoning by drugs'

(Richings et al, 1986).The high suicide rate in women doctors is of

great concern and is difficult to explain, but oneimportant factor is likely to be the higherprevalence of psychiatric morbidity in womendoctors than their male counterparts. Studies havealso shown that three-quarters of doctors who killthemselves are depressed at the time.

When students enter medical school theprevalence of psychiatric disorder is slightly, butnot significantly, higher for women than men(Miller & Surtees, 1991; Guthrie et al, 1995). Asdoctors progress in their careers, psychologicalmorbidity increases disproportionately for women.

The recent large study by Borrill et ni (1996)ofthe mental health of 12000National Health Service(NHS) employees (including doctors) found theoverall prevalence of psychological morbidity indoctors was 27%.This was using the GHQ-12 anda conservative cut-off score of three-quarters,which had been validated on a subsample of 600respondents. The prevalence of psychiatric illnessfor women doctors was 36%, which was significantly higher than that for men (24%).

Doctors are not the most psychologicallydisturbed occupational group in the NHS. Overall,health service managers and nurses reportedhigher rates of psychological morbidity thandoctors in Borrill et til's (1996) study. All three

groups, however, had significantly higher rates ofpsychological morbidity than employed workers(data from the British Household Panel Survey(BHPS), 3rd wave data, 1993-1994).

When the occupational groups in the NHS studywere compared separately for each gender, femaledoctors and female managers were found to havevery high rates of psychological morbiditycompared with men in equivalent posts. For men,there was little variation in psychological morbidity across occupations, whereas for women, theprevalence rate varied from 17% for ancillaryworkers to 36%for doctors and 41%for managers.

Box 1. Suicide risk and psychologicalmorbidity in doctors compared with thegeneral population

Suicide riskWomen doctors are 3.2 times more likely to

commit suicideMen doctors are 1.4 times more likely to

commit suicide

Psychological morbidityOverall prevalence in doctors is ~27%Prevalence in women doctors is 36%Prevalence in men doctors is 24%

The higher levels of mental ill-health recordedby women managers and doctors compared withwomen in other NHS occupations were stronglyassociated with higher levels of perceived workdemands, long working hours, working extraunpaid hours and the amount of face-to-facepatient contact (doctors only). Despite thesefindings, the researchers found it difficult toexplain the difference between men and womendoctors, as all the above factors are equally relevantfor both genders.

Rates of psychological morbidity for men andwomen are only slightly different until the agewhen women marry and have children. The ratesfor women then become much higher. Marriage isa protective factor for men in terms of psychiatricillness and a risk factor for women.

It is possible that trying to combine a professional career, which is exceptionally demandingin terms of long hours and additional commitments, with a traditional marriage and children,is particularly noxious for women. It is still therule rather than the exception that in a marriagebetween two professionals, the woman, in additionto her job, takes on a greater share of the householdcommitments. It is also still the case that manymale doctors are supported through their trainingby wives who play a subordinate role, whereas fewfemale doctors have the luxury of house-husbandsto support them.

This, however, does not explain the high ratesof suicide in female doctors as the Office ofPopulation Censuses and Surveys occupationalmortality data are based on the rates reported forsingle women. This high suicide rate in femaledoctors warrants more detailed research. Inparticular, we need to understand the complexinterplay between work demands and other

APT (1997), vol. 3, p. 278 Guthrie & Black

psychosocial factors, and we need better ways ofidentifying doctors who are at risk of suicide.

Although doctors as a profession have drasticallyreduced the amount that they smoke, there is littleevidence of a reduction in alcohol consumption.Although some studies have suggested thatdoctors may have higher than expected rates ofalcoholism (Murray, 1976; Brooke et al, 1991) theevidence is equivocal, and Clare, who reviewedthis subject in 1990, concluded that there was littleevidence to suggest that doctors drink more thanother professional groups. It is of course importantto remember that alcohol consumption amongsocial class I (professional) groups is high. Thereis also some evidence that doctors have slightlyelevated mortality rates compared with the generalpopulation for alcohol-related cirrhosis. However,other professional groups also have elevated rates,particularly lawyers.

Junior doctors

There have been three recent studies of psychological morbidity in junior doctors (Firth-Cozens,1987, 1995; Humphris et al, 1994; Grainger et al,1995). In 1986, Firth-Cozens reported a study ofpsychological distress in 318 fourth-year medicalstudents at three different medical schools. Shesubsequently followed-up these students whenthey were junior house officers (Firth-Cozens, 1987)and then later as junior hospital doctors andgeneral practitioners (GPs) (Firth-Cozens, 1995).She used the GHQ-30 with a cut-off of three-quarters and found the estimated prevalence ofemotional disturbance among the young doctorswas approximately 50%. This cut-off is lessstringent than the three-quarter cut-off used on theGHQ-12 by Borrill et al (1996) who found anestimated prevalence of 32.4% for junior hospitaldoctors in their study.

Overwork was rated as being the most stressfulfactor (on average the doctors were working morethan 90 hours per week) followed by talking todistressed relatives, serious treatment failures andeffects on the doctors' personal lives (Box 2). Firth-

Cozens found a high correlation between stressfulaspects of work and GHQ scores. There were alsohigh correlations between difficulties adjusting tochanges in sleep patterns and depression. Therewas no correlation, however, between the numberof hours worked on-call or the number of beds forwhich the doctors were responsible and levels ofdepression. Equally there was no relationshipbetween alcohol intake and GHQ scores although

Box 2 Stress and job satisfaction

fob factors which junior doctors find moststressful

Overwork; talking to distressed relatives; serious treatment failures; anddeleterious effects of work on theirprivate life

Key causes of stress for GPsOverwork; recent NHS changes and

reforms; and patients' demands and

expectations

Key components of job satisfaction inconsultants

Good relationships with patients,relatives and staff; professionalstatus; high level of autonomy; varietyof work

Job satisfaction protects doctors instressful jobs from psychologicalmorbidity

one-fifth of doctors reported occasional or frequentbouts of heavy drinking.

Firth-Cozens (1987) found highly significantcorrelations between the levels of stress reportedby junior doctors and their perception of theirwork. The more distressed the doctors were, themore negatively they saw aspects of their roles,particularly in relation to consultants. The mostdistressed doctors felt less able to discuss problemswith senior colleagues and were less sure what wasexpected of them.

As with Borrill's findings, Firth-Cozens found a

higher estimated prevalence of psychologicalmorbidity in women doctors compared with men.This contrasted with her findings when the groupwere medical students as there was no differenceat that time between the genders.

Grainger et al (1995) have recently reported astudy of 235 pre-registration house officers inBirmingham. They used the Occupational StressIndicator rather than the GHQ but found similarresults to Firth-Cozens (1987) and Borrill et al(1996). They found high rates of psychologicalillness, with women reporting worse mental andphysical health compared with men. Alcoholconsumption was high, particularly for womendoctors - twice as many women doctors weredrinking in excess of the recommended limits foralcohol consumption compared with women in thegeneral population.

Psychiatrie disorder, stress & burn-out APT (1997), vol. 3, p. 279

Nearly two-thirds of their sample regrettedstudying medicine, the main reasons cited for thisbeing long hours of work, too much responsibility,and interference with the rest of their lives. Morewomen than men experienced regret. Eighteen percent of the house officers were unsure aboutcontinuing in medicine beyond senior house officerlevel and 20 had taken steps to leave the medicalprofession.

Hospital consultants

The most comprehensive study of psychologicalmorbidity in hospital consultants is that byRamirez et ni (1996). They screened 1133 consultants in four different specialities: gastroenterology,surgery, radiology and oncology. They used theGHQ-12 with the same cut-off as Borrill et al (1996).The response rate was 78% and the overallestimated psychological morbidity for the groupwas 27%. There was no significant differencebetween the different specialities. However, only12% of the sample were women, reflecting thegender disparity in consultant appointments in thespecialities studied.

When gender is taken into account, the figuresfrom this study are very similar to those of Borrillet al, and also to the 23% prevalence of depressionand anxiety (using the Hospital Depression andAnxiety Scale) found by Caplan (1994) in his sampleof 65 consultants based in North Lincolnshire.

Of particular interest in this study (Ramirez etal, 1996) is the relationship that the researchersfound between job stress, satisfaction and mentalhealth. Doctors who reported high levels of jobstress also reported high levels of burn-out(emotional exhaustion and high depersonalisation)and psychiatric morbidity. Job satisfaction,however, appeared to protect or buffer doctorsagainst the negative relationship between job stressand burn-out and psychiatric morbidity. Doctorswho reported high stress and low satisfaction weresix times more likely to be probable cases on theGHQ than doctors who reported high stress buthigh satisfaction (see Box 2). Key components of jobsatisfaction for the doctors in this study were goodrelationships with patients, relatives and staff;professional status and self-esteem; high level ofautonomy; and variety of work (see Box 2). Surgeonsreported the highest levels of job satisfaction.

Consultants who felt insufficiently trained incommunication skills and management skillsreported high levels of depersonalisation and lowpersonal accomplishment.

Individual specialities

Some older studies have suggested that particularspecialities within medicine may be linked toespecially high rates of psychiatric disorder. Thesehave included psychiatry and anaesthetics. Arecent unpublished study by the authors of over100 psychiatrists in Manchester found very similarestimated prevalence rates of psychologicalmorbidity to the other studies quoted above,suggesting that psychiatrists do not have higherrates of psychiatric disorder than other doctors.However, rates for suicide among psychiatrists,pathologists and anaesthetists have been reportedas being higher than in some other medicalspecialities, although all the research in this areais rather dated.

Ramirez et al (1996) found that although therewas no difference in psychological morbidityamong the different consultant groups, radiologistsreported a significantly lower sense of personalsatisfaction on the MBI than gastroenterologists,surgeons or oncologists.

General practice andcommunity-based work

The largest study of mental health in GPs (Cooperet al, 1989) surveyed a sample of 4000 GPs andreceived 1817 (45.4%) replies. In contrast to otherstudies, the researchers found that women GPshad lower anxiety and depression scores thanstandardised population norms (Crown-Crisp

Experiential Index) whereas male GPs had higheranxiety scores than the normal population butsimilar scores for depression. In a second, smallerstudy (n=917), four years later (Sutherland &Cooper, 1993), and after the introduction of a newcontract of employment for GPs, the researchersfound a similar pattern of results. Women GPs hadlower anxiety scores than population norms whilemale GPs still reported higher scores thanpopulation norms (Occupational Stress Index).Overall, GPs reported lower job satisfaction thanin the previous survey, prior to the contract change.Job dissatisfaction was most strongly related to'demands of the job and increased patientexpectations'. These included increasing fear of

assault, worry about complaints from patients, andadverse publicity from the media. Other keyfactors included the conflict between clinical careand new role demands, fear of making mistakes,

APT (1997), vol. 3, p. 280 Guthrie & Black

mundane administrative work and insufficientresources. GPs reported that they no longer felt incontrol of the events that affected their ways ofworking.

Caplan (1994) surveyed 322 GPs, 81 hospitalconsultants and 121senior hospital managers. Heused the GHQ-30 and the Hospital Depression andAnxiety Scale (HAD). Scores on the HAD anxietyscale were similar among the three groups withapproximately 30%scoring >10.GPs, however, hadsignificantly higher scores for depression thanhospital managers, with consultants scoringbetween these two groups. GPs were also significantly more likely to report experiencingsuicidal ideas than hospital consultants. This is incontrast to Borrill et al (1996)who found that NHSmanagers reported the highest rates of estimatedpsychiatric morbidity. The difference in the twostudies may be due to differences in sampling anduse of questionnaires, but it is also possible that inthe past two years NHS managers have becomemore stressed. Caplan did not provide separateresults for men and women, so the influence ofgender on the overall scores cannot be assessed.

In a recent study examining stress in 532 GPs,506 hospital doctors and 241 managementconsultants and accountants, McKevitt et al (1995)found that 72%of GPs and 50%of hospital doctorsreported that their jobs were stressful most or allof the time. BothGPs and hospital doctors reportedoverwork as being a major cause of stress but GPsranked 'other people's demands upon them andexpectations upon them' much higher thanhospital doctors, and also ranked 'changes in theway they were expected to do their job' much more

highly than their hospital counterparts. Thissuggests the recent NHS reforms may be having agreater impact upon GPs than hospital doctors,although this may well change with time.

Summary

Most of the work looking at psychologicalmorbidity and stress in the medical professionmust be treated with caution as it is predominantlybased upon the use of self-report questionnaires.Several recent large-scale studies have foundremarkably consistent results, which suggest thatdoctors of all grades and specialities have higherrates of psychological morbidity than the generalpopulation. Rates of psychological morbidityare also high in other health service workers,particularly senior managers. Female hospitaldoctors have higher rates of psychological

morbidity than their male counterparts, and alsohigher rates of psychological morbidity than otherhealth service workers with the exception of seniormanagers. The situation appears to be somewhatdifferent in general practice with less evidence ofpsychological distress in women GPs. The studiesin general practice, however, have used differentmeasures than the hospital-based studies, so thefindings are not directly comparable. It is also thecase that many women GPs work on a part-timebasis, whereas part-time posts in the generalhospital setting are still relatively rare.

In the past few years there are few senior doctorsand GPs who have not experienced an increase injob demands and stress as a result of the healthservice reforms. Job satisfaction appears to protectdoctors against the noxious effects of stress uponpsychological ill health. In busy jobs, however, itis often the most enjoyable components of the job(e.g. teaching, running a special clinic or research)which are sacrificed in order to meet additionalclinical or administrative demands. Strategiesdesigned to reduce psychological morbidity indoctors will not only have to address individualcoping and managerial skills,but alsoadministrativeissues relating to the increasingwork loads ofdoctorsand increasing patient expectations.

References

British Medical Association (1992) Stress and the MedicalProfession.London: BMA.

(1993) The Morbidity and Mortality of the Medical Profession.London: BMA Board of Science and Education.

Borrill, C. S., Wall, M. A., West, G. E.,etal (1996)Mento/ Health ofthe Workforcein NHS Trusts. Final Report. Leeds: Institute ofWork Psychology, University of Sheffield and Department ofPsychology, University of Leeds.

Brooke, D, Edwards, G. & Taylor, C. (1991) Addiction as anoccupational hazard: 144 doctors with drug and alcoholproblems. British ¡ournalof Addiction, 86,1011-1016.

Caplan, R. P. (1994) Stress, anxiety, and depression in hospitalconsultants, general practitoners, and senior health servicemanagers. British Medical journal, 309, 1261-1263.

Clare, A. W. (1990) The alcohol problem in universities and theprofessions. Alcoholand Alcoholism, 25, 277-285.

Cooper, C. L., Sloan, S. J. & Williams, S. (1988)Occupational StressIndicator Guide. Windsor: NFER-Nelson.

, Rout, U. & Faragher, B. (1989) Mental health, jobsatisfaction, and job stress among general practitioners.British Medical Journal, 298, 366-370.

Crown, S. & Crisp, A. H. (1979) Manual of the Crown-CrispExperiential Index. London: Hodder and Stoughton.

Edelwich, J. & Brodsky, A. (1980) Burn-out: Stages of Disillusionment in the Helping Professions. New York: HumanSciences Press.

Firth-Cozens, J. (1987) Emotional distress in junior houseofficers. British Medical Journal, 295, 533-536.

(1995) Stress in Doctors: A Longitudinal Studi/. Leeds:University of Leeds.

Firth, J. & Morrison, L. (1986) What stresses health professionals? A coding system for their answers. British Journal ofClinical Psychology, 25,309-310.

Psychiatric disorder, stress & bum-out APT (1997), vol. 3, p. 281

Freundenberger, H. ]. & Richelson, G. (1980) Burnout: The HighCost of High Achievement. Garden City, NY: Doubleday.

Glass, D. G. & McKnight, J. D. (1996) Perceived control,depressive symptomatology, and professional burnout: areview of the evidence. Psychology and Health, 11, 23-48.

Goldberg, D. & Williams, P.A. (1988)A User's Guide to the GeneralHealth Questionnaire. Berkshire: NFER-Nelson.

Graham, J. R. (1987) TheMMPI. A Practical Guide, 2nd edn. NewYork: Oxford University Press.

Grainger, C., Harris, E., Temple, J., et al (1995) Job satisfactionand health of house offices in the West Midlands. HealthTrends, 27, 27-30.

Guthrie, E. A., Black, D., Shaw, C. M., et al (1995) Embarkingupon a medical career: psychological morbidity in first yearmedical students. Medical Education, 29, 337-341.

Humphris, G., Kaney, S., Broomfield, D., et al (1994) Stress inJunior Medical and Dental Staff. Liverpool: University ofLiverpool and Mersey Regional Health Authority.

Kelly, S., Charlton, J. & Jenkins, R. (1995) Suicide deaths inEngland and Wales, 1982-1992: the contribution of occupation and geography. Population Trends, 80,16-25.

Maslach, C. & Jackson, S. (1986)Mi7s/iïC/iBurnout Inventory (2ndedn). Palo Alto, CA: Consulting Psychologist's Press.

Miller, P. McC. & Surtees, P.C. (1991) Psychological symptomsand their course in first-year medical students as assessedby the interval general health questionnaire (I-GHQ). BritishJournal of Psychiatry, 159,199-207.

McCranie, E. W. & Brandsma, J. M. (1988) Personalityantecedents of burnout among middle-aged physicians.Behavioural Medicine, 14, 30-36.

McKevitt,C, Morgan, M., Simpson, }.,et al (1995) Doctors'Health

and Needsfor Services. London: Nuffield Provincial HospitalsTrust.

Murray, R. M. (1976) Alcoholism amongst male doctors inScotland. Lancet, ii, 729-731.

Pines, A., Aronson, E, & Kafry, D. (1981) Burnout: From Tediumto Personal Groivth. New York: Free Press.

Ramirez, A., Graham, J., Richards, M. A., et al (1996) Mentalhealth of hospital consultants: the effects of stress andsatisfaction at work. Lancet, 347, 724-728.

Richings, J. C., Khara, G. S., & McDowell, M. (1986) Suicide inyoung doctors. British Journal of Psychiatry, 149, 475-478.

Steppacher, R. C. & Mausner, J. S. (1974) Suicide in male andfemale physicians. Journal of the American Medical Association,228, 323-328.

Sutherland, V.J. & Cooper, C. L. (1992) Job stress, satisfaction,and mental health among general practitioners before andafter introduction of new contract. BritisltMedicalJournal, 304,1545-1548.

Zigmond, A. S. & Snaith, R. P. (1983) The hospital anxiety anddepression scale. Acta Psychiatrica Scandinavica, 67,361-370.

2. In junior doctors:a severity of depression is directly related to

the amount of hours that they workb over one-half are cases on the GHQc overwork is the most frequent cause of stressd stress at work is highly correlated with GHQ

scorese the prevalence of psychological morbidity is

higher in men than women.

3. In hospital consultants:a surgeons report higher stress than all other

groupsb radiologists are the most satisfied with their

jobsc insufficient training in communication skills

is associated with feelings ofdepersonalisation and low personalaccomplishment

d a key component of job satisfaction is havinga good relationship with patients

e approximately one-quarter score as cases onthe GHQ.

4. In general practitioners:a men report higher levels of anxiety than

population normsb women report higher levels of anxiety than

population normsc job dissatisfaction is strongly related to

increasing patient expectationsd job satisfaction is currently lower than it was

before the recent NHS changese overwork is a major cause of stress.

Multiple choice questions

1. In doctors:a the suicide rate is higher for men than womenb the prevalence of alcohol problems is higher

than for all other professionsc job satisfaction protects those in stressful jobs

from psychological morbidityd high levels of job stress are associated with

high levels of burn-oute the prevalence of psychological morbidity is

higher than in all other professional groupsin the NHS.

MCQanswers1abcdeFFTTF2abCdeFFTTF3abcdeFFTTT4abcdeTFTTT