astudy of neurosis and occupation - oem.bmj.com · astudy ofneurosis andoccupation 189 may be given...

Brit. J. industr. Med., 1973, 30, 187-198

A study of neurosis and occupation

DAVID FERGUSONSchool of Public Health and Tropical Medicine, The University of Sydney

Ferguson, D. (1973). British Journal ofIndustrial Medicine, 30, 187-198. A study of neurosis andoccupation. Claims that male telegraphists in an Australian communications undertakingwere unduly subject to neurosis and certain psychosomatic disorders as a result of the stress oftheir work were investigated by sickness absence and environmental and prevalence studies.The absence records of all telegraphists in the mainland capital city offices of the undertakingwere compared with those of random samples of clerks and mechanics and, because of excessabsence among Sydney telegraphists, with those of mail sorters in that city. Subsequently, 516telegraphists, 93% of those available in Sydney, Melbourne, and Brisbane, and 155 Sydneymail sorters (79% of a sample) were examined medically.Absence attributed to neurosis was much commoner in telegraphists than in the other

occupations in each capital, and in Sydney telegraphists than in those of other capitals.Employees having such absence were more likely than others also to have uncertified andrepeated absences, and absence attributed to bronchial and dyspeptic disorder and to injury.One-third (33%) of the 516 telegraphists examined were considered to have or to have haddisabling neurosis, the prevalence being much greater in Sydney (44%) than in Melbourne(19 %) or Brisbane (26 %O). The onset, course, associations, and other characteristics of neurosisare described.There was some evidence that the neurotic employee had increased liability to some other

disorders but also that he was more likely to report ill health than others. Interpretation ofincreased other ill health in neurosis is confounded by the effects of an excess indulgence inhabits. An increase in indices of mental stress was noted but some disorders commonlyattributed to stress were not unduly prevalent in neurotics. Loss of craft status, monotony,dissatisfaction with job, fear of displacement by machine, group size, and supervisorypractices were all thought to predispose to the high prevalence of neurosis in Sydney tele-graphists. However, personal and social maladjustment was particularly evident in telegra-phists in that city, and the population from which telegraphists were drawn may have beenless well adjusted in Sydney than in Melbourne or Brisbane.Though it was possible in general to characterize the employee liable to neurosis, the

predictive power of the characterization would be poor. The disorder followed no one pattern.Rather it appeared to be a collection of clinical syndromes which present as a result of thecomplex interaction of the personality with multiple factors at work and elsewhere over mostof a lifetime. In individual subjects the relationship of stress at work to symptoms was usuallyill defined, even in cases in which the identified probable factors were mainly or solely occupa-tional. Nevertheless, there seems much to be gained from the establishment of mental healthprogrammes in industry.

Claims by unions in an Australian communications 1969). Other aspects of the investigation, onundertaking that the health of their members was occupational cramp (Ferguson, 1971a), on repeatedbeing affected by the stress of their work were absences (Ferguson, 1972), and on the possibleextensively investigated by sickness absence and effects of stress on health (Ferguson, 1973), haveenvironmental and prevalence studies (Ferguson, been reported already. The present purpose is to

187

on October 12, 2019 by guest. P

rotected by copyright.http://oem

.bmj.com

/B

r J Ind Med: first published as 10.1136/oem

.30.2.187 on 1 April 1973. D

ownloaded from

http://oem.bmj.com/

188 David Ferguson

consider prevalence and associations of neurosis inthe population studied with a view to identifyingoccupational influences on neurosis.

Neurosis is not readily definable. There is nounanimity that it is a single entity rather than acollection of disorders. Diagnosis is based largely onsymptoms and syndromes (Eysenck, 1960a). Thedistinction between clinically recognizable neurosisand normal mental defence mechanisms is notsharply drawn. It is not certain that neurosis lies ona continuum of mental health (Caplan, 1964), norto what extent it is genetically or environmentallydetermined (Eysenck, 1963; Slater, 1964). Theattempt to relate an acute neurotic episode todefinite situations or objects is often a relativefailure. Ignorance of cause has bred two maincamps, the one whose followers adhere to psycho-analytical theories and the other to information andlearning theory using conditioning methods intreatment (Eysenck, 1960b). Some neurotic reactionsare accompanied by measurable biochemical change,and severe disorders of mood are now thought tohave a biochemical basis (Davies, 1969).

Studies conclusively relating neurosis to environ-mental or occupational influences are few. Brill andBeebe (1955), in a follow-up of war neuroses, foundthat though the more dangerous spheres of dutycarried higher rates of breakdown, 84% of men withovert neurosis had impaired preservice adjustment.Fraser (1947), in his wartime study of factoryworkers, considered that predisposition to neuroticillness could best be estimated from the individual'shealth record, personality, and physique. However,more neurosis than usual was found in associationwith some organizational and environmental stressesat work, including monotonous work. Of variousinfluences at work suspected of adversely affectingmental health, those often the subject of scrutinyinclude automation (World- Health Organisation,1959; Raffle, 1965; Parmeggiani, 1966; Browne,1966), shift work (Wyatt and Marriott, 1953;Thiis-Evensen, 1957; Andersen, 1958; Aanonsen,1964; Lobban, 1965; Dirken, 1966; Trumbull,1966; Taylor, 1967; McGirr, 1968; Martinez-O'Ferrall, 1968; Ferguson, 1971b), fatigue (Welford,1965a; Brown, 1967; Cameron, 1968; Ferguson,1971c), size of working group and job satisfaction(Revans, 1960; Herzberg, 1964; Welford, 1966;Mills, 1967) and noise (Broadbent, 1957; Ingham,1970). Evidence is conflicting on the effect of eachof these items on neurosis, even though they maydisturb equanimity.

Estimates of the prevalence of neurosis or of theincidence of absence attributed to neurosis varywidely according to the methods and criteria of theinvestigators (Reid, 1960; Lin and Standley, 1962).Sutherland and Whitwell (1948) found the incidenceof 'nervous disorders' in factory workers to be less

than 3 %, in marked contrast to Fraser (1947), whojudged 10% of his workers to have suffered fromdefinite and disabling neurotic illness and a further20% to have suffered from minor forms of neurosisduring the course of six months.

Various authors have associated increased absencefrom work with neurosis or neuroticism (for example,Trice, 1965; Taylor, 1968; Howell and Crown,1971; Ferguson, 1972). However, Hinkle, Plummer,and Whitney (1961) thought it wrong to assumethat all those who are frequently sick are neurotic;most neurotics work with satisfaction (Semmence,1971). Neurotics are also more likely than others tobe accident repeaters (Angeleri, Granati, andLenzi, 1964), a liability that in any one individualvaries in time with fluctuation in degree of neurosisin response to changing external stress (Smiley,1955). Neurotics have also been found more likelythan non-neurotics (Reid, 1960) or than other psy-chiatric patients (Warren, 1965) to have a previoushistory of physical ill health. To what extent theemotional stress implicit in increase of neurosisprecipitates only 'illness behaviour', not illness assuch, is still debatable (Thurlow, 1967).

Material and methodThe incidence of disabling episodes of neurosis wasestimated by a sickness absence study. The absencerecords of 765 telegraphists (the occupation mainlyunder study) continuously employed in all AustralianState capital city offices of the undertaking over atwo-and-a-half-year period were compared with those of408 clerks and 415 mechanics chosen randomly in allcapitals. Because total sickness absence was much greateramong telegraphists in Sydney than in those of any othercapital, which differed little among themselves, the 468records of telegraphists in the other capitals were com-bined for comparison with the 297 in Sydney. For thesame reason the records of 380 mail sorters and 80supervisors in Sydney were added. The method andresults in respect of absence generally (Ferguson, 1969)and repeated absence (Ferguson, 1972) have beenreported previously.

Subsequently, all telegraphists in Sydney, Melbourne,and Brisbane and the mail sorters in Sydney who wereavailable and willing (93% of telegraphists and 79% ofmail sorters) were interviewed and examined medically(Ferguson, 1971a). Sickness absence records wereavailable on those not interviewed. About half of themail sorters were interviewed by Dr. C. C. Reid as acheck on the method (Ferguson, 1969). The other half,and all of the telegraphists, were interviewed by theauthor. All groups studied were male.

Classes were selected for main groups of diseasecertified medically as causing absence, based on themethod of the London Transport Executive (1956) usingthe code numbers of the International Classification ofDiseases (World Health Organisation, 1957). Diagnosessuggestive of neurosis are often vague, and because in theeyes of some physicians and patients a stigma attaches tomental disorder a more acceptable symptomatic diagnosis



.bmj.com

/B


.30.2.187 on 1 April 1973. D

ownloaded from

http://oem.bmj.com/

A study of neurosis and occupation 189

may be given on a certificate. Codes 300 to 318 of theInternational Classification include such conditions as'disordered action of the heart' if specified as psycho-genic. Some somatic complaints symptomatic of neurosismay be missed if unspecified, but on the other handsomatic complaints attributed to neurosis, for examplenervous dyspepsia, may be wrongly included. Thirty-eightdifferent diagnoses suggestive of neurosis were noted, themost frequent being anxiety state, nervous debility, neur-osis, neurasthenia, nervous exhaustion, nervous dyspepsia,and nervous disorder, in that order. Neurosis was a fre-quent diagnosis among telegraphists in Sydney but wasrare in other capitals, and the Sydney group had a greaterdiversity of diagnoses (32) than those in the othercapitals combined (18). Nervous debility was a commondiagnosis in Sydney telegraphists but was rare in mailsorters in that city.The prevalence of neurosis was assessed from the

medical interviews and examinations without theexaminer being aware, at the time, of the absence record.A subject was not accepted as neurotic when he gave ahistory of only a transient or minimal reaction. However,a history of even one absence if severe, or persistence ofdisabling symptoms though mild, rated inclusion whenthere was no evidence of mental stress sufficient to pro-duce a reaction in an average individual. In some subjectsa history of such stress led to a diagnosis of gross stressor adult situational reaction (American PsychiatricAssociation, 1952). The rates derived for neurosis do notreflect point prevalence at the time of interview. Subjectswere neurotic if they had suffered neurosis at some timeduring their service in the undertaking to a degreeconforming to the definition by the World HealthOrganisation (1960) of a mental 'case'. Usually symptomsof neurosis even if disabling only in episodes rarelydisappeared altogether once established. The patterns ofneurosis of the American Psychiatric Association wereused as diagnostic criteria. In assessing loss of workingcapacity it was not enough that a subject should mentionabsence attributed on a certificate to neurosis; the episodeon enquiry had to be clearly of that nature. Absenceowing to neurosis was not essential for acceptance ofsymptoms as disabling if loss of social capacity waspresent to a degree which constituted taking of socialaction, as required by the World Health Organisation'sdefinition of a case. Necessarily, most cases of neurosisin a working population are slight compared with thepatient who is continuously unable to work because of thedisorder.For acceptance of neurosis the symptoms had to

conform to established patterns of the disorder and hadclearly to constitute an illness to both subject andexaminer. A solitary symptom such as sleeplessness, orvague symptoms with no recognizable pattern, or atinterview such signs as tremor, hyperreflexia, andexcessive sweating were not in themselves adequatecriteria for acceptance. Tension headache was probablya symptom of anxiety reaction though a diagnosis ofneurosis was not made on the basis of headache alone.So-called psychosomatic disorders and occupationalcramp were not accepted as evidence of neurosis.The decision whether neurosis was severe enough to

have been a disability was often difficult. Neuroticism-stability is probably a continuous dimension, and all

apparently stable persons at some time manifest symptomswhich in greater degree or persistence could be classed asneurotic. It was not possible in individual cases to gointo mechanisms of neurotic reactions or personalitydisturbances. In any case such search is often unpro-ductive or based on an unproven hypothesis. Enquiryinto genetic, constitutional, childhood, and domesticfactors was limited.

Neurosis was classed as mild, moderate or severe.Severity referred to the seriousness of the condition andwas not determined solely by ineffectiveness, to whichother characteristics contribute, nor necessarily inconformity with degree of disability. Personality andtype, degree, and duration of stress were assessed whenpossible. Degree of disability referred not to currentdisability but to disability over the period of service in theundertaking.The term 'significant' is 'used only in the statistical

sense (P <005) and the term 'associated' only if theconnection mentioned was significant. Mention of asimple association between two variables is not to denythe complex interactions inevitably shared with othervariables. Degrees of freedom of chi-square tests are notshown unless more than 1. The terms 'younger' and'older' refer respectively to employees aged less than 40,or 40 or more years.

ResultsSickness absenceOverall the proportion of employees ('neuroticabsentees') who had absence attributed to neurosis('neurotic absence') was much higher in Sydney andin telegraphists than in other capitals and occupa-tions respectively (Table 1). In every capital the ratefor telegraphists exceeded that for clerks, which inturn exceeded that for mechanics.

Despite the outstanding neurotic absentee rateamong telegraphists in Sydney, the average frequencyof neurotic absence in neurotic absentees was nogreater in this group than in other groups, and theseverity rate of neurotic absences was rather less.Neurotic absentees in Sydney mail sorters hadrelatively frequent absences of short duration.Neurotic absentee rates in Sydney telegraphists byfive-year age groups varied widely without apparentpattern. On the average, neurotic absentees in allvocations studied took two to three days moreuncertified sickness absence than others in the twoand a half years. Sydney telegraphists were notoutstanding in this regard. Younger neuroticabsentees among Sydney telegraphists had anincreased liability to absence attributed to lowerrespiratory disorder, mainly bronchitis (X2 = 4-95,P< 0-05), and to digestive disorder (X2 = 5-67,P< 002) but not to absence in any other maindiagnostic category. In every occupational group theemployee who had had absence attributed to injurywas more liable than the employee in the groupas a whole to have had neurotic absence (Fig. 1).Among Sydney telegraphists each type of repeater



.bmj.com

/B


.30.2.187 on 1 April 1973. D

ownloaded from

http://oem.bmj.com/

190 David Ferguson

TABLE 1NuMBERS OF EMPLOYEES IN CAPITAL CITY AND OCCUPATIONAL GROUPS WHOSE ABSENCE RECORDSWERE STUDIED, AND PROPORTIONS OF EMPLOYEES HAVING ABSENCE ATTRIBUTED To NEUROSIS IN 21

YEARS

Capital' Telegraphists Clerks Mechanics All occupations

n 0/n

0/ n0/n

Sydney 297 36 99 14 100 5 496 26Melbourne 192 1 1 98 9 99 0 389 8Brisbane 127 13 95 5 100 4 322 8Adelaide 79 1 1 77 5 79 1 235 6Perth 70 10 39 5 37 3 146 7

All capitals 765 21 408 8 415 3 2048'1 13

Ix2 values of tests of significance of differences between Sydney and the combined other state capitals were: for telegraphists65'6 (P< 0-001); for clerks 4-8 (P< 0-05); for mechanics 1-7 (P> 01I0); and for all occupations 102-0 (P< 0-001)."Including 380 mail sorters and 80 mail supervisors in Sydney; the rate in the former was 10%, in the latter (mostly older men)21%.

50-

~0:4 -

0iuc 30 -

-o0.=20 -

0

I)Z 0

[JAll employees

Employees with

injury absence

Sydney(263)

!Mild

Mode rate

SevereMelbourne

(136)

I

Sydney Sydney Other All All

telegraphists mail telegraphists clerks mechanicssorters

FIG. 1. Neurosis absence in employees with absence

attributed to injury and all employees in five main

occupational groups.

0 5 10Neurotic COb6)

FIG. 2. Prevalence of neurosiscapital and severity.

5s 20 25

in telegraphists, by

('chance', 'recurrent', and 'symptomatic') was morecommon in neurotic absentees than others (Fer-guson, 1972).

PrevalenceOut of 516 telegraphists interviewed in Sydney,Melbourne, and Brisbane, 171 (33 %) were regardedas having or having had disabling neurosis asdefined above. In 108 (21 %) the disorder was mild,in 53 (10%) moderate, and in 10 (2%) severe.Neurosis was commoner in Sydney (44%) than inBrisbane (26%) or Melbourne (19%) (x' = 29-01,df, P<0-001). The same order of prevalence bycapital city applied to the mild and moderate gradesof severity (Fig. 2).

Neurosis was commoner in operating telegraphists(36%) than in their (generally older) supervisoryofficers (25%.) (x2 = 5-47, P <01)2) despite thegreater overall prevalence in older (38 %) thanyounger (29%) men (x2 = 3-87, P< 0105). Thoughmild neurosis was slightly commoner in younger(22%) than older (19%) men,manymore older (19%Y)than younger (7 Y.) had moderate to severe neurosis(x2 = 15-52, P<0-001). About a quarter (24%) ofthe 86 younger and a half (49%) of the 85 older menwith neurosis ('neurotics') suffered moderately orseverely. The prevalence of neurosis by five-year agegroups (Fig. 3) agrees fairly well with the uniformage distribution noted in adults by Pasamanicket a!. (1957).



.bmj.com

/B


.30.2.187 on 1 April 1973. D

ownloaded from

http://oem.bmj.com/


60

50

40

30-

z 20

10

0Age: 20- 25- 30- 35- 40- 45- 50- S5- 60

Number: 13 77 101 100 76 18 67 40 26

FIG. 3. Prevalence of neurosis by five-year age groups in516 telegraphists.

The prevalence of neurosis in 155 older Sydneymail sorters was 15%. Each interviewer, Dr. C. C.Reid (16%) and the author (14%), found a similarprevalence in the half of this group that he examined,which supports Hamilton's (1968) contention thatobserver error in psychiatry can be reduced by carefuldefinition of criteria.

Onset and courseNo differences in prevalence of neurosis by schoolleaving age were noted. More neurotics than othersentered telegraphy after the age of 20, a differencecontributed to largely by Melbourne telegraphists(X2 = 7-29, P<0-01). In 33 cases (20%) onset ofsymptoms of neurosis occurred in childhood, and inall but 20 (12%) before the age of 40, the peakquinquennium of onset lying between 25 and 29years (23 %).

Eleven percent of subjects said they were free ofsymptoms between episodes of disabling neurosis.The great majority (92%) still had symptoms, notnecessarily disabling, at the time of interview, andin most (80%) the course was relatively stationary.In 8% the symptoms were receding, and 8% hadhad no symptoms for many years. Few (4%)reported a progressively deteriorating course. Thusthe course was mostly benign, if persistent. However,the groups studied were survivor populations; allsubjects had been continuously employed in theundertaking for at least five years before interview.Nevertheless death or invalidity retirement due toneurosis during the period of two and a half yearsin which sickness absence was studied was no morefrequent in telegraphists than in clerks, mechanicsor mail sorters. However, rates for suicide, psychosis,alcoholism, ulcer, asthma, accident,'' cancer, and'ischaemic'' or unspecified' heart disease were all

marginally (but insignificantly) greater in telegra-phists. The course of neurosis observed was generallyin keeping with the mostly favourable outcome notedfor obsessional disorder (Grimshaw, 1965) and otherforms of neurosis (Greer and Cawley, 1966), thoughit is unsatisfactory to compare people currently atwork with patients treated at psychiatric departmentsof hospitals.

Deprivation of parentsThirteen percent of 516 telegraphists had before theage of 11 been deprived of one or both parents,whether by death, divorce or other cause. Thetendency for relatively more employees with suchloss (42%) than without (32%) to be neurotic wasnot significant.

Family history of mental disorderA history of mental disorder in the immediatefamily was more than twice as common in neuroticsas in others (X2 = 22X99, P<0-001), a proportionnoted also in each capital (Table 2). With or withoutneurosis, relatively nmore in Sydney (X2 = 8-23,P< 0O01) gave a positive family history than inMelbourne or Brisbane, whose experiences werealmost identical in this respect. Neurotics who had apositive family history were also more likely thanother neurotics to admit to an unhappy, disturbedor broken adult domestic life (X2 = 6X11, P<0-02).

TABLE 2NEUROSIS AND FAMILY HISTORY OF MENTAL

DISORDER IN TELEGRAPHISTS

Family history of mental disorder %Neurosis

Sydney Melbourne Brisbane All(263) (136) (117) (516)

Present 35 28 29 33Absent .. 17 11 12 14

Drug takingRegular taking of sedatives or tranquillizers forlong periods was about as common as regular takingof analgesics, mainly aspirin or aspirin, phenacetin,and codeine compounds (Table 3). Considerabledifferences existed between capitals in the proportionsof persons taking analgesics (X2 = 24-88, P<0-001)and sedatives, explained partly by differences in agestructure between capitals but mainly by differencesin prevalence of neurosis. More younger (15 %)than older (10%) men regularly took analgesics,whereas more older (17%) than younger (9%) tooksedatives (X2 = 4.37, P< 0-05). More neurotics(25 %) than non-neurotics (7 %) took analgesics



.bmj.com

/B


.30.2.187 on 1 April 1973. D

ownloaded from

http://oem.bmj.com/

192 David Ferguson

TABLE 3TAKING OF SEDATIVES AND ANALGESICS BY

TELEGRAPHISTS

Drug Sydney Melbourne Brisbane All(263) (136) (117) (516)

______ _

Analgesic . 19 4 10 13Sedative . . 13 4 19 12

(X2 = 14-11, P<0 001) and also sedatives (29% and3 %). The difference in the case of sedatives wasprobably the result of prescription, although organicbromides could be bought over the counter. Anal-gesics would have been almost entirely unprescribed.The liability of neurotics to take drugs of either

class was associated, as would be expected, withtension headache, severity of neurosis, insomnia,irritability, depression, anxiety, and psychiatrictreatment.

Other ill health in neuroticsOver half of telegraphists with (57%) or without(51 %) neurosis had one or two other diagnoses madeat interview. Only disorders of some importance orrelevance were considered, such as asthma, chronicdyspepsia, or hypertension. One-third of either group(each 32 %) had three or more. Insignificantly fewerneurotics (12%) than others (17%) had no otherdiagnoses.

Neurosis diagnosed at interview was associatedwith frequency of certified absence over a two-and-a-half-year period preceding the interviews (Table 4).However, the difference was almost entirely con-tributed to by the experience in Brisbane andMelbourne. The diagnosis of neurosis at interviewwas not associated with the taking of uncertifiedabsence, contrary to the observation that suchabsence was increased in neurotic absentees. Half(50%) of neurotics among telegraphists in Sydney,one-third of those in Brisbane (32%), and a quarterin Melbourne (26%) had had neurotic absence.However, many non-neurotics had absence attributedto neurosis, more in Sydney (28 %) than Melbourne

TABLE 4PERCENTAGES OF TELEGRAPHISTS WITH AND WITHOUT

NEUROSIS HAVING VARIOUS NUMBERS OFCERTIFIED ABSENCES IN 21 YEARS

Numbers of absences in 2j yearsNeurosis

0-2 3-7 8+

Present (140) .. 43 38 19Absent (262) .. 57 33 10

(9%) or Brisbane (4%) (P<0-001), and more inyounger (33 %) than older (20%.) men. Amongtelegraphists in Sydney, insignificantly fewer menwith (14%) neurosis than without (22%) obtainedunjustified certificates of absence.

Symptoms of neurosisA symptom of neurosis was not accepted unless itwas characteristic of the disorder and severe enoughto have caused disquiet, discomfort or disabilityover an appreciable period. Symptoms were classedas mental (disorders of perception, thinking, affectand memory), physical (autonomic, mainly), andhabit.

Differences between younger and older subjectsin the prevalence of mental symptoms were minor(Table 5), with the exception of insomnia anddepression (both P < 0-002), which tended to occurtogether. The mean number of symptoms per subjectwas 4-6.

TABLE 5PREVALENCE OF MENTAL SYMPTOMS IN NEUROSIS

Percentages

Symptom Age 20-39 Age 40-65 Age 20-65(86) (85) (171)

Tension.. .... 72 79 75Anxiety .. .. 69 69 69Irritability .. .. 55 60 57Insomnia .. .. 38 60 49Depression .. .. 38 58 47Restlessness .. .. 45 39 43Fatigue .. .. 33 44 37Emotional lability .. 22 19 21Phobias.. .. .. 15 21 19Obsession-compulsion 16 20 18Indecision .. .. 8 14 11Weakness .. .. 12 8 10

Overall each symptom was associated with someother symptoms and other attributes of neurosis onsimple contingency tests; some 12 selected attributesout of 73 tested are arrayed in Table 6. For example,subjects who presented symptoms of depression weremore liable than other neurotics to take drugs, tohave allergies, and to be sleepless, restless, readilyfatigued, inadequate, emotionally labile, and worriedby responsibility and interpersonal domestic issues.They were also more liable to stutter, to have aseverer degree of neurosis, and to require psychiatricattention. Though noise at work was often claimedas a source of aggravation of symptoms in men withirritability and tension, it did not on the whole troublethose who were depressed, presumably because theywere withdrawn. It was often not possible todifferentiate reactive and endogenous depression.



.bmj.com

/B


.30.2.187 on 1 April 1973. D

ownloaded from

http://oem.bmj.com/


TABLE 6INTER-ASSOCIATIONS OF SELECTED ATTRIBUTES IN 115 NEUROTIC SYDNEY TELEGRAPHISTS

Attribute 1 2 3 4 5 6 7 8 9 10 11 12

I Drug taking .. .. .. * * ** * *** *

2 Insomnia .. .. .. .. ** *3 Anxiety ..4 Depression5 Sense of inadequacy6 Neurotic absence7 GP advice sought..8 Referral to psychiatrist9 Severity of neurosis10 Onset at early age11 Older age at interview12 Operator v supervisor

*P<0 05 **P<0 01 ***P<0 001

Physical symptoms in neurosis were accepted ifregarded by the subject as having resulted frommental disturbance and not from organic disease.They were recorded only if experienced under littleor no perceived stress, that is, if they were in-appropriate. The symptoms were often poorlyrelated by the subject to external stress. In respect ofall symptoms except palpitations and breathlessness,prevalence was insignificantly greater in youngerthan older subjects (Table 7). The symptoms were

TABLE 7PERCENTAGE PREVALENCE OF PHYSICAL SYMPTOMS IN

NEUROSIS

PercentageSymptom

Age 20-39 Age 40-65 Age 20-65(86) (85) (171)

Sweating .. .. 45 35 40Trembling .. .. 37 35 36Abdominal discomfort 40 29 35Palpitations .. 26 28 27Dizziness, faintness 26 19 23Diarrhoea .. 14 13 14Breathlessness .. .. 6 15 1 1

extensively interassociated. Only five subjects ad-mitted to bedwetting at any age, and 12 to tics inadult life. Nail biting (20%) and stuttering (21 %)as adults were each acknowledged by about one-fifthof neurotics. At interview excessive sweating ofpalms, axillae or face, increased tendon reflexes, andfinger tremor were all significantly more commonsigns in neurotics than in others.

Stated contributory influences in neurosisMost subjects mentioned more than one influence

as being contributory to their symptoms. Amongpersonal and domestic reasons, the commonestwere ill health of self or family (29%), housingfinance (24 %), other domestic difficulty, for example,marital discord, trouble with in-laws (29%), andsense of inadequacy (27 %). Attribution of symptomsto a sense of inadequacy was associated with indicesof severity and early onset of neurosis, whichsuggests a lifelong personality defect. Whetheradmitted as contributory to neurosis or not, 30% ofneurotics reported an unhappy home life in childhoodand 19% as an adult, the prevalence at both stagesof life being greater in Sydney than in Melbourne orBrisbane.Of stated occupational influences in neurosis,

inability to cope with the job (44 Y.), noise (37 %),monotony (31%), responsibility (18 %), job dis-satisfaction (16%), supervisory inadequacy (14%),and poor ventilation (14%) were the commonestmentioned by subjects. Inability to cope, noise, andresponsibility were each much more often mentionedby older than younger subjects.On the basis of all available information in each

case it was concluded that 26% of neurotics more orless exclusively experienced symptoms at work orrelated their occurrence thereto and that in 10%symptoms arose away from work. In 64 %, symptomscould not be related solely to one or other environ-ment. Rating of apparent external stress, using theclassification of the American Psychiatric Association(1952), revealed evidence in only 22% of cases of adegree of stress sufficient to permit some causalrelationship with neurosis to be reasonably estab-lished (Ferguson, 1973).

ImpairmentThe degree of impairment in neurosis generally ranparallel with severity. Of the 171 neurotic subjects,



.bmj.com

/B


.30.2.187 on 1 April 1973. D

ownloaded from

http://oem.bmj.com/

194 David Ferguson

35% were judged to have minimal impairment, 43%mild (definitely impaired social or vocationaladjustment), 21 % moderate (seriously impairedadjustment), and 1 % severe. Severe impairmentmostly would have precluded continued employ-ment. In general, degree of impairment increasedmarkedly with age.

Other associations of neurosisSignificant associations (mostly P<0 001) werenoted between neurosis generally and 17 otherattributes. Some associations, such as those withtension headache, sweating, hyperreflexia, andtremor, are part of the symptom complex of neurosis.The greater susceptibility of neurotics to annoyanceby noise, noted by Broadbent (1957), was confirmed.Neurotics preferred day shift to night, despiteapparently greater stresses in day work. Adverseattitudes to job and supervision were stronglyevident. The association of occupational cramp withneurosis is well established (Ferguson, 1971a).Neurosis was also linked with dyspepsia, pepticulcer, smoking, and drinking, all of which wereinterassociated. The neurotic, significantly moreoften than other telegraphists, rented accommoda-tion, indulged in neither sport nor hobby, and hadspondylosis and recurrent myalgia of the trunk.On the average the grip strength of neurotics wasrelatively weak.

Associations with some attributes did not quitereach the 5% level of significance. Neurotics tendedto be brought up in the city (noted also for NewYorkers by Srole et al. (1962)), to live more than anhour's travel from work, to be deprived of one orboth parents before the age of 11 years, to be of lessthan average height, and to avoid double jobbing.No tendency (P> 010) was found for neurosis to beassociated with other disorders, for example, hyper-tension, obesity, and asthma, sometimes claimed tobe causally related thereto or to mental stress.The predictive attribute analysis of Macnaughton-

Smith (Williams and Lance, 1968) failed to showhigh-order interactions that were helpful in pre-dicting neurosis. Application of the programmeMULTIST (Lance and Williams, 1967) to 72qualitative attributes, 11 numerical attributes, and9 multi-state attributes revealed no high-leveldiscontinuity and suggested that the system hadlittle structure. A more powerful 'flexible' sortingprocedure showed a division of neurotics into threesubgroups; however, the types overlapped, and thecollections of characteristics were viewed as peaks ina continuum.

Comparisons with Sydney telegraphistsTelegraphists in Sydney in general comparedunfavourably with their colleagues in Melbourneand Brisbane in respect of some attributes relevant

to neurosis. The Sydney men aged 30 to 39 years hadsignificantly more often remained single or had abroken marriage, and older telegraphists had moreoften married after the age of 30 years. Sydneytelegraphists were more likely to enter the occupa-tion after the age of 20. Smoking, drinking, andnon-narcotic analgesic taking were much moreprevalent in Sydney, as also were 'unjustified'certified sickness absence and a family history ofmental disorder. Telegraphists in Sydney were alsomuch more likely than those in Melbourne andBrisbane to express dissatisfaction with their job,with supervision, and with opportunity for advance-ment. Younger men more often had a sense ofinsecurity in their job.Compared with older Sydney mail sorters, older

Sydney telegraphists significantly more often took asecond job, found work unsatisfying and opportunityinadequate, were dissatisfied with supervision andannoyed by work noise, drank moderately toheavily, and smoked more than 14 cigarettes daily.Though the greater prevalence of drinking andsmoking in telegraphists in Sydney than in Melbourneor Brisbane could possibly be attributed to socialdifferences between cities, the differences in the habitsbetween vocations in Sydney suggests that occupa-tion affected their prevalence. Despite factorslessening the comparability of older members of thetwo vocations in Sydney, the difference in prevalenceof neurosis was great enough (49 to 15%) toindicate that influences at work probably contributedmore than domestic or other social influences.By implication, the greater prevalence in telegraphistsin Sydney than in Melbourne or Brisbane was alsowork-caused. The only probable occupationalinfluences on neurosis revealed by comparison of thetwo vocations in Sydney were those reflected in theadverse attitudes to job satisfaction, opportunity,and supervision. Though annoyance by noise waslinked with neurosis, exposure to noise was similarin each capital. More double jobbing in Sydneytelegraphists than in mail sorters cannot be adducedas a factor in neurosis because a negative associationbetween these attributes was found to exist intelegraphists; and second jobs were no more pre-valent in Sydney, where neurosis was common, thanin Melbourne where it was relatively uncommon.Other differences between the two Sydney groupswhich may have indicated vocational variation insocial conditions were negligible or at least notsignificant. In matters such as conjugal state, num-bers of children, deprivation of parents, homeownership, urban-rural origin, participation insport and social activities, unjusified absence, andanalgesic taking, mail sorters fared no better norworse than telegraphists, suggesting that theseattributes did not contribute particularly to neurosisin the latter.



.bmj.com

/B


.30.2.187 on 1 April 1973. D

ownloaded from

http://oem.bmj.com/


Discussion

In each mainland state capital office of the under-taking studied, absence attributed to neurosis wascommoner in telegraphists than in the other voca-tions studied, and in any vocation in Sydney thanin the same vocation in other capitals. Thus, occupa-tional and geographic differences existed, the com-bination of the two influences producing aparticularly adverse experience in Sydney telegra-phists. That the differences were real was confirmedby the prevalence study. Even in the absence ofdisabling neurosis, many employees displayed anunsatisfactory personal, social, and work adjustmentthat fell far short of optimal mental health. Whetheras a result or not, many indulged in habits to anextent detrimental to physical health.On Hogerzeil's (1968) hypothesis the absence

pattern in Sydney telegraphists suggested a responseto unfavourable management practices coupled withmental stresses exceeding capacity for ready adjust-ment.The neurotic had more absence attributed to

sickness than the non-neurotic and an increase ofcertain physical disorders, though not significantlymore diagnoses of other chronic ill health made atinterview. Thus, limited confirmation was given ofaccounts that neurotics are more likely than stablepersons to suffer other disorders (Reid, 1960), butalso of the tendency of neurotics to report rather thanto have ill health. There was some evidence inneurotics connecting smoking and drinking withulcer and dyspepsia but evidence also of increasedoccupational and other mental stress. However,diseases sometimes ascribed at least in part to stress,such asessential hypertension, coronary heart disease,bronchial asthma, and migraine, were not par-ticularly prevalent in neurotics. Occupationalstress appeared to loom larger than non-occupationalstress, contrary to the findings of Kbppich et al.(1968) in women.

In some respects the characteristics of the neuroticin the present study confirmed associations noted byFraser (1947), for example, with muscular weakness,other illness, boredom, dislike of job, sedentarywork, and disturbed human relationships outsidework. The results also confirmed some findings ofKornhauser (1965) that mental health at work variesdirectly with job level and satisfaction and manage-ment practices, and inversely with the broadness ofthe hierarchical base.

Physical environmental conditions at workprobably exerted any adverse effects throughcontribution to the general dissatisfaction evidentin the offices. Lighting and the visual task werestressful in various ways, yet were subject of littlecomplaint by neurotics, in contrast to Fraser's(1947) findings. Surroundings were drab, amenities

far from ideal, and ventilatory control often poor.Uncomfortable seating, crowding of work positions,and layout ofequipment may have aggravatedoperat-ing difficulty. Overall noise levels of about 83dBA would certainly have interfered with com-munication and possibly with performance, and didexcite behavioural response particularly in neurotics.However, in general the physical environment wasof a similar standard in the three capital officesinspected, and thus could not be adduced as reasonfor the excess neurosis in Sydney telegraphists.The semi-automated system of telegraphs, intro-

duced in 1958, was probably a source of stress, lessfrom its operation than from its implications forstatus, satisfaction, security, and opportunity. Thechange brought role alterations involving loss of bothhierarchical and craft status for many. It brought alsomonotony, loss of personal contact, and machinepacing with less opportunity for job satisfaction. Itbrought unjustified fear of displacement by womenor machine, yet men were frustrated at being trappedin a vocation the skills for which had little applicationelsewhere.The effect of size may have operated at several

levels. City size, noted by Chiesman (1957) to affectabsence attributed to neurotic and digestive disordersin postmen, was probably not a factor in sicknessabsence in the present study. Apart from Sydney'spoor record (Sydney is the largest capital inAustralia), offices in other mainland capitals, whichvary widely in size, varied little in incidence ofabsence. Size of undertaking, noted by Revans (1960)to affect absence and by Kornhauser (1965) to reflectmental health indices, probably had an effect ondifferences in neurosis between capitals, or at leastcreated conditions of group 'non-support' whichfoster tension and antagonism (McClintock, 1965).Workroom size may have influenced neuroticabsence through its effect on small group structureand on supervisory effectiveness (Welford, 1965b).To Bashford (1942) and Smith (1956) supervision wasa main determinant of neurotic absence, but theapparent supervisory failure in Sydney was toogreat to attribute to size alone. The mental ill healthimplicit in adverse attitudes was to some extent ajustified reaction to an intolerable situation (Frenchand Kahn, 1962).Some personal and social attributes apparently

had little of their postulated stressful effect, at leastas measured by the prevalence of neurosis andadverse attitudes. The strongest and most numerousassociations with mental and physical health werenoted in respect ofsmoking and drinking, themselvesstrongly linked. The very strong associations ofthese drug habits with neurosis probably reflectedcharacteristics of the neurotic personality. How-ever, the habits may themselves create anxiety, orcontribute to general ill health.



.bmj.com

/B


.30.2.187 on 1 April 1973. D

ownloaded from

http://oem.bmj.com/

196 David Ferguson

There is no obvious reason why entrants totelegraphy should be particularly predisposed toneurosis (Smith, Culpin, and Farmer, 1927). Toassume that they were would be to presuppose that agroup of unstable persons was drawn on, whereasmost men moved into telegraphy as a normal lineof advancement in the undertaking. Within tele-graphy the prevalence of neurosis differed betweenSydney and the other two capitals studied, yet thestandards of recruitment were said to be the samethroughout Australia. The population from whichentrants were drawn should not have differed inpredisposition to neurosis, suggesting that conditionsof service were responsible for making any pre-disposition manifest. However, personal and socialmaladjustment, for example in marital state, habits,and taking of unjustified absence, was more evidentamong the telegraphists in Sydney than in Melbourneor Brisbane. To what extent this observation indi-cated reaction or selection (Knupfer, Clark, andRoom, 1966) is uncertain.There are indications of relative social mal-

adjustment generally in New South Wales, whosecapital Sydney is. For example, rates of divorce,litigation, and drinking are higher than in otherstates. Entry to telegraphy occurred at a later age inSydney. A high staff supervisory ratio was anunfavourable recruitment factor in that city.Selection of the less apt may thus have occurred, inaddition to any unfavourable factors at work, whichis in keeping with the hypothesis of recruitment ofthe less healthy to less favoured occupations (Lin andStandley, 1962).The influence of stress on mental health is often

overemphasized (Atkin, 1962). Mental reactionstend to be fitted to current theories of mentalmechanisms often despite the patient's rejection ofthe link. Though stress acts on the predisposed, sothat a stress is a function of the individual and of theenvironment, the occurrence of postulated signs ofgroup occupational stress in an undertaking, such asneurosis, absenteeism, resentment of authority, anddrinking (Ferguson, 1972; 1973), are as muchwarnings of failure in the undertaking as in theindividual. Stress is not necessarily harmful; ifsupport is available, events leading to mental stressmay be strengthening, not weakening (Caplan, 1964).

Despite the ill-defined nature of the relationshipbetween occupational stress and neurosis, thereseems much to be gained from establishing a mentalhealth programme in an undertaking. Some effortshould be made to select applicants suited to theirintended placement. Prediction of future neurosisis at present unsurely based on tests of personality.Exclusion from stressful tasks of all who recordscores suggestive of emotional instability would keepout many who would be quite successful and whowould have no mental disability. Exclusion would

be on surer ground if based in addition on a previouspoor work and sickness record and history of frankemotional disturbance in the applicant, and onpersonality disorder, conflict or separation inparents.The more effective use of persons with marginally

unstable personalities could be combined withelimination of the more emotionally stressful aspectsof the job. The organization of work may bemodified to reverse the unfavourable attitudes foundto be strongly associated with neurosis. Emotionalsecurity and morale may be improved by job enlarge-ment, by increased identification, involvement, andcommunication, and by removal of irksome andunnecessary administrative procedures (Revans,1960). The size effect may be lessened by decentraliz-ing authority, increasing initiative and respon-sibility, and decreasing administrative rigidity.The present study has revealed a relative failure of

social adjustments in Sydney telegraphists comparedto those in Melbourne and Brisbane. In consideringoccupational health, the two environments, workand home, cannot be divorced. A mental healthprogramme in an undertaking should include somefacility for counselling and health education,preferably centred on an effective occupationalhealth service. An assessment of social adjustmentand emotional stability could become part of thepreplacement and periodic medical examination, sothat the personality resources of the employee couldbe matched with the demands of the job; mentaldifficulty could then be foreshadowed, and thephysician would be in a position to advise in theearly stages of neurosis.

This study was carried out under the general direction ofDr. Gordon C. Smith, Head of the Occupational HealthSection of the School of Public Health and TropicalMedicine, to whom I am indebted for helpful criticism.I am grateful also to the Computer Section of the Com-monwealth Scientific and Industrial Research Organiza-tion, particularly to Mr C. H. Gray and Dr. W. T.Williams, for some of the data handling, and to Dr. C. C.Reid for helping with interviews of the control group.

References

Aanonsen, A. (1964). Shift Work and Health. NorwegianMonographs on Medical Science. Oslo.

American Psychiatric Association (1952). Diagnostic andStatistical Manual. Mental Disorders. Washington.

Andersen, E. J. (1953). The main results of the Danishmedico-psychosocial investigation of shift workers. Pro-ceedings of the XII International Congress on OccutpationalHealth 1957, Vol III, Helsinki, pp. 135-136.

Angeleri, F., Granati, A., and Lenzi, R. (1964). Nevrosi elavoro femminile nell 'industria. Folia Medica, 47, 157-185.

Atkin, 1. (1962). Mental disorders and external stress. Lancet,1, 581-582.

Bashford, H. (1942). Supervision of sick absence in industry.Lancet, 2, 289-290.



.bmj.com

/B


.30.2.187 on 1 April 1973. D

ownloaded from

http://oem.bmj.com/


Brill, N. Q., and Beebe, G. W. (1955). A Follow-up Study ofWar Neuroses. Veterans Administration Medical Mono-graph. Washington.

Broadbent, D. E. (1957). Effects of noise on behavior. InHandbook of Noise Control, edited by C. M. Harris,Chapter 10. McGraw-Hill, New York.

Brown, I. D. (1967). Measurement of fatigue in abnormalphysical environments. In The Effects of AbnormalPhysical Conditions at Work, edited by C. N. Davies, P. R.Davis, and F. H. Tyrer, pp. 168-178. Livingstone,Edinburgh.

Browne, R. C. (1966). Automation and stress. Journal ofPsychosomatic Research, 10, 73-75.

Cameron, C. (1968). A questionnaire study of fatigue in civilaircrew. Human Engineering Note 24. AeronauticalResearch Laboratories, Melbourne.

Caplan, G. (1964). Principles of Preventive Psvchiatry. Basic,New York.

Chiesman, W. E. (1957). Absenteeism in industry. Clinicalaspects of absenteeism, Royal Society of Health Journal,77, 681-686.

Davies. B. (1969). Recent studies of severe depressive illnesses.Part 2. Medical Journal of Australia, 1, 557-565.

Dirken, J. M. (1966). Industrial shift-work: decrease inwellbeing and specific effects. Ergonomics, 9, 115-124.

Eysenck, H. J. (1960a). Classification and the problem ofdiagnosis. In Handbook of Abnormal Psychology, editedby H. J. Eysenck, pp. 1-31. Pitman, London.

(I 960b). Behaviour Therapy and the Neuroses. Pergamon,Oxford.

(1963). Biological basis of personality. Nature, 199,1031-1034.

Ferguson, D. (1969). Stress and Health in Telegraphy. MDthesis, University of Sydney.

(1971a). An Australian study of telegraphists' cramp.British Journal of Industrial Medicine, 28, 280-285.

(1971b). Shift work and health. Personnel PracticeBulletin, 27, 113-122.

(1971c). Fatigue and the organisation of work. InIntroduction to Ergonomics, pp. 126-132. ProductivityPromotion Council of Australia.

(1972). Some characteristics of repeated sicknessabsence. British Journal of Industrial Medicine, 29, 420-431.

(1973). A study of occupational stress and health.Ergonomics (in press).

Fraser, R. (1947). The incidence of neurosis among factoryworkers. MRC Industrial Health Research Board Report,No. 90. H.M.S.O., London.

French, J. R. P., and Kahn, R. L. (1962). A programmaticapproach to studying the industrial environment andmental health. Journal ofSocial Issues, 18, No. 3, pp. 1-47.

Greer, H. S., and Cawley, R. H. (1966). Natural Historyof Neurotic Illness. Australasian Medical Association,Sydney.

Grimshaw, L. (1965). The outcome of obsessional disorder.British Journal ofPsychiatry, 111, 1051-1055.

Hamilton, M. (1968). Observer error in psychiatry. Pro-ceedings of the Royal Society of Medicine, 61, 453-454.

Herzberg, F. (1964). The motivation-hygiene concept andproblems of manpower. Personncl Administration, 27,3-7.

Hinkle, L. E., Plummer, N., and Whitney, L. H. (1961). Thecontinuity of patterns of illness and the prediction offuture health. Journal of Occupational Medicine, 3, 417-423.

Hogerzeil, H. H. W. (1968). The effects on sickness absence ofpersonal characteristics. In Proceedings of the Symposiumon Absence from Work Attributed to Sickness, edited byA. W. Gardner, pp. 55-56. Society of OccupationalMedicine, London.

Howell, R. W., and Crown, S. (1971). Sickness absence levels

and peisonality inventory scores. British Journal ofIndustrial Medicine, 28, 126-130.

Ingham, J. C. (1970). Symptoms of psychological disorder innoisy environments. In Proceedings of Symposium on thePsychological Effects of Noise, edited by W. Taylor.University of Dundee. Society of Occupational Medicine,London.

Knupfer, G., Clark, W., and Room, R. (1966). The mentalhealth of the unmarried. American Journal of Psychiatry.122, 841-851.

Koppich, A., Preda, N., Gavrilescu, N., Elias, R., andVoicu, Gh. (1968). Beitrage zum Studium der Atiologieder asthenischen Neurose bei Arbeiterinnen mit vor-wiegend nervoser Beanspruchung. Internationales Archivfdr Gewerbepathologie und Gewerbehygiene, 24, 221-225.

Kornhauser, A. (1965). Mental Health of the IndustrialWorker. John Wiley, New York.

Lance, G. N., and Williams, W. T. (1967). Mixed-dataclassificatory programs 1. Agglomerative systems.Australian Computer Journal, 1, 15-20.

Lin, T., and Standlky, C. C. (1962). The Scope ofEpidemiologyin Psychiatry. W.H.O., Geneva.

Lobban, M. C. (1965). Time, light and diurnal rhythms. InThe Physiology of Human Survival, edited by 0. G.Edholm and A. L. Bacharach, pp. 351-386. AcademicPress, London.

London Transport Executive (1956). Health in Industry.A Contribution to the Study of Sickness Absence. Butter-worth, London.

McClintock, C. G. (1965). Group support, satisfaction, andthe behaviour profiles of group members. British Journalof Social and Clinical Psychology, 4, 169-174.

McGirr, P. 0. M. (1968). Circadian rhythms in flight.Transactions of the Society of Occupational Medicine, 18,3-12.

Martinez-O'Ferrall, J. A. (1968). Circadian rhythms. Journatof Occupational Medicine, 10, 305-315.

Mills, J. D. (1967). Job satisfaction in large factories.Personnel Practice Bulletin, 23, 252-260.

Parmeggiani, L. (1966). Automation and OccupationalHealth. I.L.O., Geneva.

Pasamanick, B., Roberts, D. W., Lemkau, P. V., and Krueger,D. E. (1957). A survey of mental diseases in an urbanpopulation. American Journal of Public Health, 47, 923-929.

Raffle, P. A. B. (1965). Human factors in automation.Transactions of the Association of Industrial MedicalOfficers, 15, 88-95.

Reid, D. D. (1960). Epidemiological Methods in the Study ofMental Disorders. WHO, Genzva.

Revans, R. W. (1960). Morale and the size of the workinggroup. In Modern Trends in Occupational Health, editedby R. S. F. Schilling. Butterworth, London.

Semmence, A. (1971). Rising sickness absence in GreatBritain-A general practitioner's view. Journal of theRoyal College of General Practitioners, 21, 125-246.

Slater, E. (1964). Genetical factors in neurosis. BritishJournal of Psychology, 55, 265-269.

Smiley, J. A. (1955). A clinical study of a group of accident-prone workers. British Journal of Industrial Medicine, 12,263-278.

Smith, J. A. (1956). Occupational stress and emotional illness.Journal of the American Medical Association, 161, 1038-1040.

Smith, M., Culpin, M. and Farmer, E. (1927). A study oftelegraphists' cramp. MRC Industrial Fatigue ResearchBoard Report No. 43. H.M.S.O., London.

Srole, L., Langner, T. S., Michael, S. T., Opler, M. K., andRennie, T. A. C. (1962). Mental Health in the Metropolis.The Midtown Manhattan Study. McGraw-Hill, NewYork.

Sutherland, I., and Whitwell, G. P. B. (1948). Studies in



.bmj.com

/B


.30.2.187 on 1 April 1973. D

ownloaded from

http://oem.bmj.com/

198 David Ferguson

occupational morbidity (2). British Journal of IndustrialMedicine, 5, 77-87.

Taylor, P. J. (1967). Individual variations in sickness absence.British Journal of Industrial Medicine, 24, 169-177.

(1968). Personal factors associated with sicknessabsence. British Journal ofIndustrial Medicine, 25, 106-118.

Thiis-Evensen, E. (1957). Shift work and health. In Reportsof the XII International Congress on Occupational Health,1957, Helsinki, vol. 1, pp. 97-105.

Thurlow, H. J. (1967). General susceptibility to illness.Canadian Medical Association Journal, 97, 1397-1404.

Trice, H. M. (1965). Alcoholic employees: a comparison ofpsychotic, neurotic and 'normal' personnel. Journal ofOccupational Medicine, 7, 94-99.

Trumbull, R. (1966). Diurnal cycles and work-rest schedulingin unusual environments. Human Factors, 8, 385-398.

Warren, W. (1965). A study of adolescent psychiatric in-patients and the outcome six or more years later. 1.Clinical histories and hospital findings. Journal of ChildPsychology and Psychiatry and Allied Disciplines, 6, 1-17.

Welford, A. T. (1965a). Fatigue and monotony. In The

Physiology of Human Survival, edited by 0. G. Edholmand A. L. Bacharach, pp. 431-463. Academic Press,London.

- (1965b). Stress and achievement. Australian Journal ofPsychology, 17, 1-11.

- (1966). Individual capacity and social demands. InOperational Research and the Social Sciences, edited byJ. R. Lawrence, pp. 531-542. Tavistock, London.

Williams, W. T., and Lance, G. N. (1968). Choice of strategyin the analysis of complex data. The Statistician, 18,31-43.

World Health Organisation (1957). International Classifica-tion of Diseases. W.H.O., Geneva.

(1959). Mental Health Problems of Automation.W.H.O., Geneva.

- (1960). Epidemiology of Mental Disorders. TechnicalReport Series No. 185. W.H.O., Geneva.

Wyatt, S., and Marriott, R. (1953). Night work and shiftchange. British Journal ofIndustrial Medicine, 10, 164-172.

Received for publication July 18, 1972.Accepted for publication November 14,1972.



.bmj.com

/B


.30.2.187 on 1 April 1973. D

ownloaded from

http://oem.bmj.com/

astudy of neurosis and occupation - oem.bmj.com · astudy ofneurosis andoccupation 189 may be given...

Documents