Stress at work and mental health status among female ... › content › oemed › 47 › 1 › 20.full.pdfEstryn-Behar,Kaminski,Peigne,Bonnet, Vaichere,Gozlan,Azoulay,Giorgi Anindex

British Journal of Industrial Medicine 1990;47:20-28

Stress at work and mental health status among femalehospital workers

M Estryn-Behar, M Kaminski, E Peigne, N Bonnet, E Vaichere, C Gozlan, S Azoulay,M Giorgi

AbstractRelations between working conditions andmental health status of female hospital work-ers were studied in a sample of 1505 women:43% were nurses, 32% auxiliaries, and 7%ancillary staff; 13% were other qualified healthcare staff, mainly head nurses; 5% had occupa-tions other than direct health care; 63%worked on the morning, 20% on the afternoon,and 17% on the night shift. Data were collectedat the annual routine medical visit by theoccupational health practitioner, using selfadministered questionnaires and clinicalassessments. Five health indicators were con-sidered: a high score to the general healthquestionnaire (GHQ); fatigue; sleep impair-ment; use of antidepressants, sleeping pills, orsedatives; and diagnosis of psychiatric mor-bidity at clinical assessment. Four indices ofstress at work were defined: job stress, mentalload, insufficiency in internal training anddiscussion, and strain caused by schedule. Theanalysis was conducted by multiple logisticregression, controlling for type of occupation,shift, number ofyears ofwork in hospital, dailytravel time to work, age, marital status, num-ber of children, and wish to move house. Sleepimpairment was mostly linked to shift andstrain due to schedule. For all other indicatorsof mental health impairment and especiallyhigh GHQ scores, the adjusted odds ratiosincreased significantly with the levels of jobstress, mental load, and strain due to schedule.This evidence of association between workinvolving an excessive cumulation of stressfactors and mental wellbeing should be con-sidered in interventions aimed at improvingthe working conditions of hospital workers.

Assistance Publique, Mission Etudes et InformationSante-Travail, Ergonomie H6tel-Dieu, 75004 Paris,FranceM Estryn-Behar, E PeigneINSERM U 149,94807 Villejuif CedexM KaminskiAssistance Publique, M6decine du Travail, ParisN Bonnet, E Vaichere, C Gozlan, S Azoulay, M Giorgi

Since 1960, the use of medical time for anxiety andneurotic reactional states and psychogenic depres-sion has been increasing.' Work itself is not usuallyconsidered as the sole cause of psychic disorders butit may decompensate vulnerable states of psychicstability. The part of occupational life events in theoccurrence of depression has received recentinterest.' Protecting mental wellbeing at workrequires that general stressors be identified and theproportion of workers suffering from psychologicalor somatic stress reactions be known in order todevelop adequate corrective and preventive actions.5The work ofwomen in hospital is a privileged field

for studying the respective roles of working condi-tions and sociodemographic factors in the prevalenceof mental and sleep disorders. These womenrepresent nearly 9% of salaried female workers inFrance.6 Recent studies have shown that workers inoccupations involving intense social contactsexperience symptoms related to the loss of mentalresources, known as the "bum out phenomenon."This phenomenon has often been described in healthcare occupations but only among one occupationalgroup, nurses, and mainly in intensive care orpaediatric units.' Descriptive studies of thefrequency ofwork stressors among hospital staffhavebeen published'° and assessments of mental wellbeing using the general health questionnaire(GHC)"" have been made among nurses'3 andmedical students."4 Comprehensive epidemiologicalstudies analysing the relations between work stres-sors, sociodemographic factors, and stress levelamong health care personnel in various occupationalcategories have yet to be developed; and this is theaim of the present paper.

Population and methodsSAMPLE AND DATA COLLECTIONThe present study is part of a general study ofworking conditions and health status of femalehospital workers, which associates an epi-demiological and an ergonomic approach. Theprotocol has been described in detail elsewhere (MEstryn-Behar, unpublished data); the main charac-teristics are summarised below.The study was carried out in 12 public hospitals in

the Paris area, all belonging to the same regional

20

on June 15, 2021 by guest. Protected by copyright.

http://oem.bm

j.com/

Br J Ind M

ed: first published as 10.1136/oem.47.1.20 on 1 January 1990. D

ownloaded from

http://oem.bmj.com/

Stress at work and mental health status amongfemale hospital workers

board; 26 departments were selected to cover alltypes of specialties except psychiatry. For eachspecialty, two departments were selected in twodifferent hospitals. The sample considered for studyincluded all female workers in post in these depart-ments on 1 January 1986, except physicians. Datawere collected at the annual routine medical visitwith the hospital staff occupational health physician.Physicians are not covered by this surveillancesystem and thus could not be included in the study.Data on occupation and working conditions,

demographic and social characteristics, health status,treatments, and sick leave in the preceding 12 monthswere collected by a self administered questionnaire.Health status was also assessed by a clinical examina-tion. Supplementary data on sick leave were obtainedfrom administrative sources. The participation ratewas 90% and data were finally collected for 1505women. For the 10% ofwomen missing, some basicinformation was available, including occupation andduration of sick leave.Among the 1505 women, 21% worked in

paediatric or maternity departments, 30% in medicaldepartments, 29% in surgical departments, 11% inintensive care units, and 9% in geriatric depart-ments.Ergonomic studies were conducted in 10 depart-

ments.'5 In each department one nurse in each shift-(morning, afternoon, and night) and one nursing aideon the morning or afternoon shift was followed upduring the whole period. Interruptions and reorgan-isations of the work programme were recorded;discussions with patients, other members of staff,and doctors were recorded, as well as their duration;ambiguities, errors, and search for information couldbe observed in relation to work organisation, ways oftransmitting information, and types of leadership;time spent in patients' rooms was measured.

Semistructured interviews of the workers wereconducted at the end ofeach ergonomic study and therecords were analysed by two psychiatrists.

HEALTH INDICATORSPsychological disorders may take insidious formssuch as tiredness, dizziness, headache, sleepproblems, sexual troubles, aggressiveness, and psy-chosomatic disorders. After the stage of fatigue anddissatisfaction, the subject may present a majordepressive disorder. In our context of primary care,to screen for unrecognised or early states of diseasesthat might develop into more important disorders,we used the following five indicators.Fatigue-If the woman said she was waking up

tired, if she was presently particularly tired, and if, inthe previous 12 months, she had carried on workingdespite exhaustion, or if she had given at least two ofthese answers (32% of the women).

Sleep impairment-When the woman slept less

than six hours on workdays, had regular sleepproblems on workdays, and had regular sleepproblems on rest days, or if she described at least twoof these situations (31%).

Use of drugs such as antidepressants, sedatives, orsleeping pills (28%).

Psychiatric morbidity assessed by the occupationalhealth physician at the routine medical visit (mostlydepressive state or irritability) (21%).

The GHQ was used in its 12 item version.'2 Eachitem was rated on a four point scale, with weightsfrom 0 to 3. Scores higher than 12 were considered asan indicator of impaired mental wellbeing, and werefound in 26% of the sample. Previous studies havevalidated the use of the GHQ in occupational com-munities12 16 and in the present study the GHQ scorewas found to be highly correlated with the othermental health indicators (table 1): each was two tothree times more frequent when the GHQ scoreswere high. A similar relation was found betweenGHQ scores and duration of absence from work forsickleave in the previous 12 months (table 1).The GHQ score has also been analysed quan-

titatively (mean value 10-5); the results were similarto those obtained for the dichotomised variable butare not shown in this paper.

OCCUPATION AND WORKING CONDITIONSThe women were grouped into five occupationalcategories: 43% were nurses; 32% nursing aides; 7%ancillary staff; 13% were other qualified health carestaff, mainly head nurses, but also midwives,physiotherapists, psychologists, and dieticians; and5% had occupation other than direct health care(secretaries, laboratory technicians, social workers).

Sixty three per cent of women worked in themorning shift (beginiing between 0600 and 0900),20% in the aftemoon shift (beginning between 1200and 1530), and 17% in the night shift (beginningbetween 1800 and 2400).Four indices of stress at work were defined to

summarise the main characteristics of workingconditions that may affect mental health status (seeappendix for description ofindividual items includedin the indices):

Table I Mental health indicators according to GHQ score*

GHQ score GHQ score< 12 > 12(No) % (No) % p Value

Fatigue (1049) 23 (362) 56 p

Estryn-Behar, Kaminski, Peigne, Bonnet, Vaichere, Gozlan, Azoulay, Giorgi

An index of job stress, scoring 0 to 8, includingadequacy between training and actual tasks, relationswith patients, interest in work, and satisfaction; 64%ofwomen were at the lower level of job stress (score0-1), 28% at the intermediate level (score 2-3), and8% at the higher level (score 4-8).An index of mental load, in three levels, including

interruptions in tasks, need of frequent reorganisa-tion of daily work programme, and overwork: 18% ofthe sample were at the lower level, 62% at theintermediate level, and 20% at the higher level.An index of insufficient internal training and discus-

sion, scoring 0 to 5, including teaching within thedepartment, explanations about tasks, discussions ofproblems about relations with patients or organisa-tion, and progress in job; this index had a lower level(score 0-1) for 59% of women, an intermediate level(score 2-3) for 35%, and a higher level (score 4-5) for6%.An index of strain caused by schedule, scoring 0 to

3, including number ofconsecutive working days anddissatisfaction with working hours; 13% of womenwere at the higher level (score 2-3).Three other factors related to work were also

considered: number of years since beginning work inhospital, daily time travelling between home andworkplace, and an index of physical load of workincluding posture and heavy lifting.

SOCIODEMOGRAPHIC FACTORSFactors known to be related to mental health wereincluded as potential confounders: age, maritalstatus, number ofchildren at home, and wish to movehouse.217 15

ANALYSISThe association between the health indicators andoccupation, working conditions, and indices of jobstress and sociodemographic factors was analysed,firstly, by crude comparisons, using Pearson's chi-

squared test and, secondly, by multiple logisticregression. Adjusted odds ratios and 95% confidenceintervals for the odds ratios were calculated foreach occupational characteristic. Although similaranalyses were conducted on the five mental healthindicators, in some tables only fatigue, sleep impair-ment, and GHQ scores are shown. Because ofmissing data, numbers may vary according to thestudied variable.

ResultsSick leave for mental and sleep problems caused 3080days of sick leave among a total of 25 433 days. Thiscause was the third. after musculoskeletal disordersand problems during pregnancy, and before res-piratory and ENT diseases.

Fatigue and sleep impairment were more frequentamong nursing aides and ancillary staff than in theother categories (table 2). There was no significantdifference between occupational categories in the useof drugs or psychiatric morbidity at examination andno difference in GHQ scores. The only associationobserved with shift was a higher proportion of sleepimpairments among night workers (table 2).Although the level of stress varied greatly accord-

ing to occupation (table 3), high levels were notalways found in the same occupations. Job stress wasespecially high among ancillary staff and to a lesserdegree among nursing aides, whereas mental loadwas high for nurses, other qualified health care staff,and other occupations. Insufficiency in internaltraining and discussion was mentioned morefrequently by less qualified health care staff and bythose in occupations other than health care. Straindue to schedule was found mostly in nurses andancillary staff. Job stress and insufficiency in trainingand discussion did not vary according to the shift; thenight shift was characterised by a significantly lowermental load and the afternoon shift by a high level ofstrain due to schedule (results not shown). All stress

Table 2 Mental health indicators according to occupational category and shift*

PsychiatricSleep Use of morbidity

Fatigue impairment drugs at exam GHQ score(No) % (No) % (No) % (No) % >12

Occupation:Nurses (639) 29 (612) 29 (639) 23 (621) 22 (607) 27Nursing aides (479) 39 (455) 38 (479) 28 (472) 18 (444) 23Ancillary staff (96) 44 (89) 38 (96) 31 (94) 28 (87) 28Other health care staff (185) 22 (175) 27 (186) 27 (180) 21 (176) 25Other occupations (83) 31 (78) 13 (82) 22 (81) 20 (81) 28p Value p < 0001 p < 0-001 NS NS NS

Shift:Morning (923) 32 (874) 27 (922) 25 (903) 20 (872) 25Afternoon (299) 32 (280) 31 (299) 26 (291) 19 (276) 25Night (?52) 32 (245) 51 (253) 28 (246) 25 (238) 29p Value NS p < 0-001 NS NS NS

*See definitions in text.

22


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/


Table 3 Indices of stress at work (o%) according to occupational category* (number in parentheses)

Nursing Ancillary OtherOccupation Nurses aides staff health care Other p Value

Shift:Morning 52 64 72 70 100Afternoon 25 19 19 16 0 p


Table 5 Mental health indicators according to indices of stress at work

PsychiatricSleep Use of morbidity GHQ score

Fatigue impairment drugs at exam (> 12)(No) % (No) % (No) % (No) % (No) %

Job stress:0-1 (928) 25 (875) 29 (926) 23 (901) 16 (874) 162-3 (409) 40 (399) 32 (410) 24 (402) 25 (388) 354-8 (122) 58 (118) 44 (122) 45 (120) 43 (118) 64p Value p


Table 8 Mental health indicators according to occupation, shift, and stress at work: adjusted odds ratios*

Fatigue Sleep impairment GHQ score (> 12)

ORt ORt ORt(n=1260) 95% CIt (n= 1217) 95% CIt (n=1203) 95% CIt

Occupation:Nurses 1-0 1-0 1-0Nursing aides 1-6 1-1-2-1 1-4 1 0-1 9 0-6 0 4-0-9Auxiliary staff 1-6 0-9-2-8 1.9 1-1-3-4 0-6 0-3-1-2Other health care 0-8 0-5-1-3 0-8 0-5-1-3 1-3 0-8-2-0Other 1-5 0-8-2-7 0 5 0-2-1-2 1-5 0-8-2-8

Shift:Morning 1-0 1-0 1-0Aftemoon 1-2 08-1 7 1-5 1-0-2-1 09 0-6-1-4Night 1-4 1-0-2-1 3-8 2-6-5-5 1-6 1-1-2-5

Job stress:0-1 1-0 1-0 1-02-3 1-6 1-2-2-2 1.0 0-7-1-3 2-4 1-8-3-44-8 29 1*8-4-7 1-5 0-9-2-4 69 4-2-11*3

Mental load:Low 1-0 1-0 1-0Average 15 1-0-21 1 1 0-7-1-5 2-1 1-3-3-3High 2-6 1-6-40 1-2 08-1 9 2-9 1-7-4-9

Insufficiency in internaltraining:0-1 1-0 1-0 1-02-3 1.1 0-8-1*4 1-2 0-9-1*7 1-7 0 8-1*54-5 1*2 0-7-2-1 1*2 0-7-2-1 1*7 0-9-2-9

Strain due to schedule:0-1 1-0 1-0 1-02-3 1*4 1*2-2 4 1*6 1*1-2-4 2-2 1*4-3 2

*Adjusted for age, marital status, number of children at home, wish to move house, number of years of work in hospital, and daily traveltime in a multiple logistic regression.tAdjusted odds ratio.t95% Confidence interval for the adjusted odds ratio.

sick leave (34 days). Reasons for non-participationwere difficulties in examining women on the nightshift (31 %) and long sick leave (420%)-mostly due toproblems during pregnancy (18%). The occupa-tional levels of participants and non-participants didnot differ significantly. The high response ratecombined with the anonymity guaranteed to allrespondents gives no reason for rejecting the resultson the grounds of reliability. As in any cross sectionalstudy some possible bias has to be considered. Selfselection out of work of those exposed to the moststressful situations, leading to a healthy workereffect, may have minimised some relations betweenhealth and working conditions; however, theobserved relations are strong. The answers on work-ing conditions may have been influenced by themental health status at the time of survey, thusoverestimating certain associations. Questions aboutworking conditions, however, were asked first in thequestionnaire before any health parameter.Some of the work stress indices were validated

through an ergonomic study. At group level theergonomic observations gave descriptions of thework that were consistent with that observed fromthe questionnaires.'5 The time spent with patientsdid not differ between nursing aides and nurses butthe basic knowledge of nursing aides was not thesame and their discussions with physicians weremore exceptional, thus explaining their higher level

of job stress and of insufficiency of internal trainingand discussion. The number of interruptions duringtasks was higher in day observations than in nightshifts, but the higher number of patients per nurseduring night shift is consistent with a lower level ofmental load only for the aftemoon shift.

Five mental health indicators were considered inorder to approach, so far as possible, the presence orabsence of psychological stress. The consistency ofthe relations between working conditions and the fivemental health indicators increased the viability ofthedata collected, overcoming semeiological problemsof diagnosis when made by primary care practition-ers and not psychiatrists. Especially for fatigue wemust question the possibility of an inverse relation:the fatigue leading to a negative judgment on workinstead of adverse working conditions leading tofatigue. But the observation of relations betweenfatigue and insufficiency of internal training anddiscussion, which is much less subjective, reinforcedthe relations between fatigue and job stress. Thereliability of the mental health indicators is reinfor-ced by the analysis by psychiatrists of the semistruc-tured interviews after the ergonomic observations:they concluded that the content of the interviewssuggested a dysphoric state of mind, linked toisolation of the staff members and lack of reinfor-cement for the tasks performed.'9 In our study overallfatigue was noted by 32% of the women. The

25


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/


morbidity at medical examination was higher thanthe one observed by a national study of morbidityamong French active women. In Sweden Joenpeltoand Vahanen found general fatigue among 16% ofhealth workers.2' Burn out is described by Maslachand Pines as a highly insidious process with a threephase sequence in the first of which health careworkers feel an emotional exhaustion, resulting fromprolonged intensive involvement with patients, inwhich just thinking of having to go to work makesthem tired.2' For Cherniss, this first stage ofimbalance between resources and demand producesstrain in the worker, manifested as feelings of ten-sion, irritability, and fatigue.22 It is necessary to act atthis phase to avoid entering in a cycle of failuredifficult to break.

Strain due to schedule and schedule itself of oursample could be compared with that of a nationalsample,of salaried women in a survey using similarquestions (MF Christofari, unpublished data); thecomparison showed a higher frequency of work,more than 100 nights a year (16% instead of 2%), ofwork early in the morning or late in the afternoon(respectively 38% instead of 12% and 20% instead of7%), of consecutive working days without rest (35%instead of 66% worked five days or fewer at astretch). By contrast, job motivation seemed to behigh among health care staff, as only 3% found theirjob without interest whereas 22% of the nationalsample found their job repetitive. In this nationalsurvey the subsample ofwomen working in hospitalsdiffered from the other salaried women on theseworking conditions. Women working in the hospitalsbelonging to the regional board studied, however,were exposed to a higher level of strain due toschedule but had a higher job motivation. The straincaused by schedule and night work necessitate also aspecific attention.2324 Chronobiological knowledgemust be integrated in organisation of worktime toavoid performance impairments such as thosestudied by Poulton et al.2"The positive impact of counselling and discussion

between staffmembers and with physicians has beenfocused in case studies by numerous authors.2"'8Some ofthem have discussed the role of staffsupportgroups.'" Others have exposed the benefit of groupmeetings with an outsider trained in psychology.Y'Fawzy et al showed that the involvement of thepsychiatric liaison in four services led to high jobsatisfaction, task orientation, and reduced turnovercompared with a department without similarsupport.37The head nurse can develop an atmosphere in

which staff members are encouraged to identifystress factors in the environment and learn frommistakes. The head nurse can use report times todiscuss psychological issues relative to patients.'839The simple provision of specific workshops, refre-

sher courses, and establishment of consensus onprotocols and techniques in use in the department isof great use to alleviate anxiety.4' Bishop has insistedon the necessity that criteria for treatment or non-treatment should be clearly discussed.4' Other auth-ors have proposed preceptorships to help newgraduates to acquire a role model and a resourceperson within the clinical setting to help new staffmembers to adjust to their jobs and also experiencednurses moving into a different specialty.42-' Vachonet al trained a staff member to be a resource personfor all the staff about psychological problems."Leppanen and Olkinuora have reviewed the

epidemiological studies on stress problems amongScandinavian health care personnel.'0 Insufficiencyin training was also found: Gardel et al showed thatnearly half of the nurse's aides thought that theirmedical knowledge was insufficient for their workdemands.' Every fourth practical nurse, every fifthregistered nurse, every tenth head nurse, and everyseventh physician considered themselves to beincompetent in medical questions; incompetenceconcerning psychological issues was experiencedeven more often: in particular the practical nursesand nurse's aides who spend more time with patientsthan other professionals thought themselves to beincompetent in psychological issues.

Several studies reviewed by Leppanen and Olkin-uora deal with role conflicts and role ambiguity'0:Joenpelto and Vahanen showed that nearly halfofthesurgical ward nurses reported that they wereunfamiliar with some of their duties.2' They alsothought that the conflicting expectations anddemands of patients and other staffmembers causedthem strain; Jokinen and Poyhonen noted that onlyone out offive practical nurses in Helsinki was alwaysaware oftasks to be done and the aims oftheir work.47Meeting the medical and emotional needs of ill

patients is a major source of stress but there are othercauses. The hospital workers are involved in a chainof command that makes them a likely target for therelease offrustrations by physicians, supervisors, andco-workers, as well as patients and their families. Butthere are few opportunities for them to deal withthese frustrations. Ifthe health workers are experien-cing personal problems unrelated to their work thestress is exacerbated. Improved communication andlearning to manage the stresses are likely to result in abetter interpersonal climate in which patients'emotional comfort could be nurtured. The medicaldirector of the department and the head nurse mustdevote time to improving communication, as is themain wish ofthose interviewed'9 and, maybe, changesome of the rules of the doctor-nurse game as statedby Thomstad et al.'

In the present state of knowledge furtheraetiological studies, primarily cohort studies, areneeded to understand better the role of work and

26


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/


work stresses in mental health. Intervention studiesare even more necessary, to evaluate the effectivenessof the measures discussed above, to occupations.

We thank C Berthier, M F Delaporte, A Lefevre,J M Leroux, M Louet, M F Maillard, G Melonio,M C Paoli, Mrs C Blanpain, N Carlin, and M Gayetfor their collaboration and Mrs S Closon, GGamarre, C Girier-Desportes, D Larcher, and MSomveille for technical support. The research waspartly supported by a grant from the InstitutNational de la Sante et de la Recherche Medicale(contrat de recherche AP No 85801 1).

AppendixINDEX OF JOB STRESS: number of the following itemsDoes your job correspond to your level?seldom or never (13%)Do you do work for which you are not trained?always (7%)Have you enough time to talk to patients?not enough or very insufficient (23%)Are you well prepared to answer their questions?often doubt or worry about answer (24%)Are you satisfied with atmosphere in presentdepartment? No (13%)Do you find your job interesting?variable or not interesting (31 %)In your opinion, atmosphere in your department is?rather bad (8%)Has your work an effect on your mood outside?frequently rather bad (19%)

INDEX OF MENTAL LOAD:combination of the followingitemsAre you interrupted in your tasks?No-rarely (13%)Sometimes (38%)Often (49%)Have you to reorganise your work programmemany times during the day?No-rarely (40%)Often (51%)Always (9%)Do you feel overworked in your job?No (23%)A little (55%)Yes (22%)

INDEX OF INSUFFICIENCY IN INTERNAL TRAINING ANDDISCUSSION: number of the following itemsAre teaching sessions organised in your department?Never (68%)Do you get explanations about tasks which you areasked to do?Not at all (22%)Are there opportunities to discuss problems related

to organisation?No (14%)Are there opportunities to discuss relations orpsychological difficulties with patients?No (26%)Do you feel you are progressing in your job?Not at all or regressing (20%)

INDEX OF STRAIN CAUSED BY SCHEDULE: number of thefollowing itemsWork more than six consecutive working days

(41%)Has asked or would ask to begin work earlier orlater (18%)Is not satisfied with schedule (10%)

1 Pichot P, Boyer P, Zarifian E, et al. La maladie depressive. Paris:Laboratoires Ciba-Ceigy, 1983.

2 Brown GW, Davidson S, Harris T, et al. Psychiatric disorders inLondon and North West. Soc Sci Med 1977;11:367-77.

3 Lloyd C. Life events and depressive disorders reviewed. ArchGen Psychiatry 1980;37:529-48.

4 Ferreri M, Vacher J, Alby JM. Facteurs evenementiels etdepressions: le questionnaire EVE. Une nouvelle approche demethodologie predictive. Psychologie Medicale 1987;19:2441-8.

5 Kalimo R, El-Batawi MA, Cooper C. Psychosocialfactors at workand their relation to health. Geneva: World Health Organisa-tion, 1987.

6 Institut National de la Statistique et des Etudes Economiques.Etude particuliere a partir de l'enquete emploi. Paris: INSEE,1988. (Serie D 128.)

7 Jacobson S. Stressful situations for neonatal intensive carenurses. American Journal of Maternal Child Nursing1978;3: 144-50.

8 Estryn-Behar M, Mascolo I. Psychopathologie du travail desprofessions paramedicales dans un service de reanimationmedicale. Psychologie medicale 1983;15:289-94.

9 Appelbaum SH. Stress management for health care professionals.Rockville, Maryland: Asper Publishers, 1981.

10 Leppanen RA, Olkinuora MA. Psychological stress experiencedby health care personnel. Scand J Work Environ Health1987;13:1-8.

11 Goldberg D. The detection ofpsychiatric illness by questionnaire.London: Oxford University Press, 1972.

12 Banks MH, Clegg CW, Jackson PR, Kemp NJ, Stafford EM,Wall TD. The use of the general health questionnaire as anindicator of mental health in occupational studies. Journal ofOccupational Psychology 1980;53:187-94.

13 Vachon MLS, Lyall WA, Freeman SHF. Measurement andmanagement of stress in health professional working withadvanced cancer patients. In: Death education. Washington:Hemisphere, 1978:365-75.

14 Firth J. Levels and sources of stress in medical students. Br MedJ 1986;292:1177-80.

15 Estryn-Behar M, Kaminski M, Peigne E, Fouillot JP, Debray Q.Concevoir une architecture et une organisation pour la sante etla securite a l'hopital. In: Proceedings of seminar on buildingforpeople in hospital: workers and consumers. Dublin: EuropeanFoundation for the Improvement of Living and WorkingConditions (in press).

16 Rahman M, Sen AK. Effect of job satisfaction on stress,performance and health in self-paced repetitive work. Int ArchOccup Environ Health 1987;59:115-21.

17 Weissman MM, Myers JK. Rates and risks of depressivesymptoms in a United States urban community. Acta Psy-chiatr Scand 1978;57:219-31.

18 Hirschfeld RMA, Cross CK. Epidemiology of affective disor-ders psychological risk factors. Arch Gen Psychiatry1982;39:35-46.

19 Debray Q, Estryn-Behar M, Guillibert E, Azoulay S, Bonnet N.Les conditions de travail du personnel hospitalier feninin: lesaspects psychologiques. Psychiatrie et Psychobiologie1988;3:389-99.

20 Joenpelto L, Vanhanen K. In: Leppanen and Olkinuora.'°

27


http://oem.bm

j.com/

Br J Ind M


ownloaded from

http://oem.bmj.com/

Estryn-Behar, Kaminski, Peigne, Bonnet, Vaichere, Gozian, Azoulay, Giorgi

21 Maslach C, Pines A. The burn out syndrome in the day caresetting. Child Care Quarterly 1977;6:100-13.

22 Cherniss C. Staff burn out: job stress in the human services.London: Sage Publications Beverly Hills, 1980.

23 Estryn-Behar M. Capacites de travail, memoire et rythmescircadiens. Bilan des etudes realisees dans le domaine dutravail de nuit des personnels hospitaliers. Sante Publique,numero special "Memoire et memoires en Sante Publique"1989;4:105-13.

24 Direction du personnel, CHSCT Central, CHSCT Locaux.Etude sur le travail de nuit. Paris: Assistance PubliqueH6pitaux de Paris, 1988.

25 Poulton EC, Hunt GM, Carpenter A, Edward RS. The perfor-mance of junior hospital doctors following reduced sleep andlong hours of work. Ergonomics 1978;21:279-95.

26 Rossini LA, Howell MC, Todres ID, Dorman J. Groupmeetings in a pediatric intensive care unit. Pediatrics1974;53:371-4.

27 Joly JB, Richard P, Baudet MT. Chronique de pediatrie sociale.Les relations entre parents-enfants et personnel soignant dansun centre de neonatologie. Pediatrie 1976;XXXI:603-13.

28 Strain JJ. Psychological reactions to acute medical illness andcritical care. Crit Care Med 1978;6:39-44.

29 MacGrath HM. Mediating stress through support networks. In:Jacobson SF, MacGrath HM, eds. Nurses under stress. NewYork: John Wiley, 1983:194-208.

30 Gonin 0. L'agressivite dans les equipes soignantes. Kranken-pflege/Soins Infirmiers 1984;4:68-9.

31 PolettiR. Quepeut-onfairedepositifavecl'agression?Krankenp-flege/Soins Infirmiers 1984;4:61-2.

32 Verspieren P. Face a celui qui meurt. Paris: Desclee de Brower,1984.

33 Zumbrunnen R. Quand un patient peut-il se montrer agressif etque peut l'equipe soignante a l'h6pital general? Krankenpflege/Soins Infirmiers 1984;4:71-3.

34 Price TR, Bergen BJ. The relationship to death as a source ofstress for nurses on a coronary care unit. Omega 1977;8:229-38.

35 Sebag-Lanoe R, Virguie Cl, Cazas A. Le travail en equipe engeriatrie: un apprentissage indispensable pour le mndecin.Medecine et Hygiene 1982;40:1842-6.

36 Zittoun R. Une communaute pour en parler. In: Entretiens avecE. Hirsch. Partir. L'accompagnement des mourants. Paris: Cerf,1986:1-14.

37 Fawzy IF, Wellisch DK, Pasnau RO, Leibowitz B. Preventingnursing burn-out: a challenge for liaison psychiatry. GeneralHospital Psychiatry 1983;5:141-9.

38 KornfeldDS. Psychiatric viewofthe intensive care unit. Br MedJ 1969;i:108-10.

39 Simon NM, Whitely S. Psychiatric consultation with MICUnurses: the consultation conference as a working group. HeartLung 1977;6:497-504.

40 Bilodeau CB. The nurse and her reactions to critical-carenursing. Heart Lung 1973;2:358-63.

41 Bishop V. Stress in the intensive care unit. Occupational Health1983;35:537-43.

42 Dell MS, Griffith E. A preceptor programme for nurses' clinicalorientation. J Nurs Adm 1977;7:37-8.

43 Chickerella BG, Lutz WJ. Professional nurturance: perceptor-ships for undergraduate nursing students. Am J Nurs1981;1:107-9.

44 Murphy ML, Hammerstad SM. Preparing a staff nurse forprecepting. Nurse Educator 1981; Sept-Oct: 17-20.

45 Jacobson SF, MacGrath HM. Nurses under stress. New York:John Wiley, 1983.

46 Gardell B, Gustafason RA, Brandt C, Tillstr6m I, Torbioun IIn: Leppanen and Olkinuora.10

47 Jokinen M, Poyhonen T. In: Leppanen and Olkinuora.'048 Thomstad B, Cunningham N, Kaplan BH. Changing the rules of

the doctor-nurse game. Nurs Outlook 1975;23:422-7.

Accepted 13 February 1989

Destruction of manuscripts

From 1 July 1985 articles submitted for publicationwill not be returned. Authors whose papers arerejected will be advised of the decision and themanuscripts will be kept under security for threemonths to deal with any inquiries and then destroyed.

28


http://oem.bm

j.com/

Br J Ind M


ownloaded from
http://oem.bmj.com/

Documents

Stress at work and mental health status among female ... › content › oemed › 47 › 1 › 20.full.pdfEstryn-Behar,Kaminski,Peigne,Bonnet, Vaichere,Gozlan,Azoulay,Giorgi Anindex