Keywords
|
| Stress, salivary cortisol, JCQ, DASS |
Introduction
|
| Healthcare staff can suffer from job stress as a result of organizational factors and an imbalance of job demands, skills and social support at work place1. Many studies had shown that healthcare staff has high prevalence of stress2-5. Increased stress has been associated with heavy workload and its effect of home life3, pressure of work, complaints from patients and relatives, risk of violence, poor management style4, inadequate resources, insufficient training, lack of control over work, low involvement in decisionmaking 5, low job satisfaction and harassment2. |
| Several studies had shown job stress has correlation with several salivary biomarkers6,7 such as salivary cortisol, salivary alpha amylase and chromogranin A. Activation of the hypothalamic–pituitary–adrenal (HPA) axis which measured by salivary cortisol reflects the chronic stress where as activation of the sympatho–adreno–medullary (SAM) system, which measured by salivary alphaamylase and chromogranin A reflects acute stress exposure8. The use of saliva as a diagnostic fluid has many advantages. It is non-invasive compared with phlebotomy, does not have the risk of needle-stick injuries and able to minimize stress compared with other invasive procedures such as phlebotomy. This will also alleviate the stress from fear of needles during collection, which may bias the results9. Process for obtaining saliva is simple and easy, self-collection after instruction is possible and there is no need for special trained staff for its collection 10. |
| Assistant medical officers are the first liners in handling patient in health clinics, hospital accident and emergency units, and thus, stress is one of the factors that could affect their quality of work and patient relationship. Hence, the objectives of this study were to determine the prevalence of stress, to determine the relationship between salivary cortisol and stress and to compare the salivary cortisol between stressed and non-stressed assistant medical officers in Ministry of Health hospitals. |
Methods
|
Study design
|
| A cross-sectional study was conducted as it was the best design to determine the relationship of the variables of interest at one particular time. Data collection was conducted from July till December 2008 on assistant medical officers in all Ministry of Health hospitals. |
Data collection procedure
|
| A list of assistant medical officers working in all departments in Ministry of Health hospitals were obtained from State Health Department. Assistant medical officers in-charge for each department were contacted via phone to obtain appropriate particulars with regards to the selection of subjects. All assistant medical officers who were randomly selected were recruited for this study. During visits to the hospitals, the researcher obtained permission from the hospital directors and briefed the assistant medical officers regarding the survey. Confidentiality of the information was assured. Self-administered questionnaire were distributed. Assistant medical officers attending both shift and non-shift work were included in this study. Those who were on prolonged oral steroid or diagnosed to have any psychiatrics illnesses or chronic medical illnesses were excluded. For salivary cortisol level, information and steps on the collection methods were explained. Morning saliva was collected between 8.00 am and 10.00 am from each assistant medical officer using Salimetrics Oral Swab (Salimetrics LLC, State College, PA, USA) by keeping it under the tongue for one to two minutes. The specimens were sent in a cool box with ice packs at 4 to 8oC to Immunology Laboratory for storage at -20oC before underwent cortisol immunoassay test using Expanded Range High Sensitivity Salivary Cortisol Enzyme Immunoassay Kit (Salimetrics LLC, State College, PA, USA). |
| Sample size was calculated based on single proportion formula11 using proportion of stress 34.0% among secondary school teachers in Kota Bharu12. Taking the precision of 0.07 at 95% confidence interval, the minimum required sample was 176. The precision was set at 0.07 after considering its clinical importance and feasibility of the study. After considering the non-response rate of 10%, the sample size calculated was 194. |
Research tools
|
| Data collection was conducted using questionnaire which consisted of three parts socio-demographic data and Depression, Anxiety and Stress Scale (DASS-42). The DASS-42 is a set of three self-reported scales designed to measure the negative emotional states of depression, anxiety and stress. Each of the three DASS scales contains 14 items, divided into subscales of 2-5 items with similar content. The 42 DASS items are presented to subjects in random order and scored on a Likert scale of 0 to 3 (0=Did not apply to me at all, 1=Applied to me to some degree, or some of the time, 2=Applied to me to a considerable degree, or a good part of the time, and 3=Applied to me very much, or most of the time) to rate the extent to which they have experienced each state over the past week 13. However, in this study, only Stress Scale of DASS-42 questionnaire will be used. |
| Stress scale consists of items 1, 6, 8, 11, 12, 14, 18, 22, 27, 29, 32, 33, 35, 39. Scores for DASS-Stress were calculated by summing the scores for relevant items and converting these scores into percentile scores which were divided into five levels of severity (normal=0- 14, mild=15–18, moderate=19–25, severe=26–33, extremely severe=34+)13. In this study, subjects who scored in the ‘normal’ range on stress were referred as non-stressed, while subjects who scored in the ‘mild’ to ‘extremely severe’ ranges were referred to as stressed14. |
| Expanded Range High Sensitivity Salivary Cortisol Enzyme Immunoassay Kit (Salimetrics LLC, State College, PA, USA) was used for salivary cortisol quantitative assessment. It is a competitive immunoassay kit specifically designed for the quantitative measurement of salivary cortisol. It is not intended for use with serum/plasma or for diagnostic use. It is intended only for research use with saliva15. |
Statistical analysis
|
| Data was entered using SPSS 12.0.1 (SPSS Inc, Chicago IL, USA) and analyzed using STATA 8.0 (StataCorp, College Station, TX, USA) for analyses. Descriptive analysis was done to obtain the frequency and percentage. Correlation analysis was done to determine the relationship between salivary cortisol and stress score. Normal distribution of Y in any point of X and equal variance of Y in any point of X assumptions were met if the scatter plot between salivary cortisol and stress score shows an “elliptical shape”. If the assumptions were met, proceed with Pearson correlation analysis. If the assumptions were not met, proceed with Spearman P correlation analysis. Correlation coefficient (r) >0.75 suggests very good to perfect correlation, 0.50 -0.75 suggests moderate to good correlation, 0.25-0.50 suggests fair correlation, and <0.25 suggests little or no correlation. Independent t test analysis was done to compare the salivary cortisol between stressed and non-stressed assistant medical officers. P value <0.05 was used as a cut-off for statistical significance. |
Results
|
| There were 194 assistant medical officers in the government hospitals this study. However, 175 have responded making a response rate of 90.7%. Those who did not respond have not completed the questionnaire. Posthoc power calculation using linear regression was 91.4%. |
| There were 24 respondents classified as stressed and hence, the prevalence of stress among assistant medical officers was 13.7% (95% CI: 8.61, 18.79). The sub-analysis of percentages of stress according to the DASS severity ratings were shown in Table 1. Table 2 showed the socio-demographic characteristics and Table 3 showed the job characteristics of 24 stressed and 151 nonstressed assistant medical officers. |
| Scatter plot between salivary cortisol and stress score indicated that normal distribution of Y in any point of X and equal variance of Y in any point of X assumptions were not met. Hence, Spearman P correlation analysis was done. There was no significant (P=0.393) correlation between salivary cortisol and stress score. The observed Spearman P correlation was 0.066 which suggest no correlation. Salivary cortisol was significantly higher (P=0.033) among stressed compared to nonstressed assistant medical officers (0.78 versus 0.67 μg/dL respectively) (Table 4). The salivary costisol of stressed assistant medical officers was 0.11 μg/dL (95% CI: 0.218, 0.01) higher compared to non-stressed assistant medical officers. |
Discussion
|
| The prevalence of stress among assistant medical officers was 13.7%. The prevalence in the current study was lower compared to study done by Azlihanis (2007) using similar research tools. In her study, she found that the prevalence of stress among secondary school teachers in Kota Bharu, Kelantan was 33.9%. The prevalence of stress was also lower compared to study done by Harmy (2001), among other group of health care workers. In his study, he found the prevalence of job strain was 23.7% among nurses in government teaching hospital in Kelantan16. Though they are different subgroups of health care workers, they have the similarity in providing care to patients. |
| One study done on occupational stress, found that women appear to experience significantly higher level of job stress compared to men17. This was supported by Aziah et al. (2004) where the proportion of high job strain among male was lower compared to female laboratory technicians in MOH and USM (36.8% versus 63.2%, 35.7% versus 64.3% respectively)18. In Taiwan, 17.2% of nurses working in psychiatric institutions had selfperceived stress19. |
| In Japan for example, male occupational physician had 0.37 times at odds (95%CI: 0.19, 0.70) of job stress compared to females20. This was supported by another study which found that male general practitioners had less job stress compared to females3. Stress was positively correlated with health problems for both women and men, but perceived stress were strongly correlated with women21. Hence, being a male, was one of the probable reasons for low prevalence of stress in this study. |
| However, Michael et al. (2009) found that sex differences in stress do not stem from a genetic or biological difference, but from social and psychological differences which were associated with age, marital status and education17. Hence, it was the gender differences, which was a social construct that may have a more significant effect on stress compared to sex, which was biologically determined. |
| Another possible reason for the lower prevalence of stress was age. Although not significant, the average age for stressed assistant medical officers was younger compared to those non-stressed (35.0 years versus 37.8 years). This was supported by a study done in Japan where younger (20 to 39 years) occupational physicians had more stress compared to those aged more than 59 years. As age had a significant positive correlation with tenure or duration of employment, this reason was considered reasonable, as younger occupational physicians may lacked of practical skills which usually acquired with tenure 20. |
| The differences in the prevalence were not directly comparable because of different periods of time, methods of data collection and reference values. DASS was used to assess stress based on score scale but JCQ was used to assess job strain based on the median cut-off of the JCQ value 22. |
| There was no significant correlation between salivary cortisol and stress score. The observed Spearman correlation was 0.066 suggest of no correlation. The non-significant finding in the correlation was probably due to the low prevalence of stress among assistant medical officers. However, salivary cortisol was significantly higher (P=0.033) among stressed compared to non-stressed assistant medical officers (0.78 versus 0.67 μg/dL respectively). This was supported by Kirschbaum et al. (1995) where he found high cortisol response among healthy men who were exposed to repeated psychological stress. |
| Elevated level of salivary cortisol among stressed assistant medical officers was probably due to their commitment in providing best care to the patients, as care giving behaviour may reflects the caregivers’ cortisol levels 7. |
| In contrast, study done among nurses in emergency department in Singapore revealed that although they perceived their job was stressful compared to general ward nurses, their salivary cortisol was lowered compared to latter 23. Study on self-perceived stress among patients attending breast cancer clinic too, unable to show correlation between stress and salivary cortisol 24. |
| Study done among middle aged men and women on workdays and weekends to determine the association between cigarette smoking and salivary cortisol found that cigarette smoking was associated with elevated cortisol level which was probably due to nicotine exposure 25. In the current study, smoking was not controlled in determining the relationship between salivary cortisol and stress score. This is because, the assumptions for linear regression between the two variables were not met and, hence, the linear regression analysis between salivary cortisol and stress score cannot be done. |
| One of the factors that could affect the level of salivary cortisol was the diurnal variation of salivary cortisol. Physiologically, salivary cortisol secretion, similar to serum cortisol, displays marked circadian rhythm, characterised by low levels towards evening, steady increase during early morning and peaking just after awakening, levels then decline rapidly9. However if a person was stressed, adrenocorticotropic hormone (ACTH) and the cortisol secretion will be elevated regardless of the time of day. Stress overrides the circadian rhythm as well as the negative feedback mechanism 26. |
| Decreased daytime cortisol decline was associated with higher level of stress in male. The decreased daytime cortisol decline indicates the hypathalamo-pituitary-adrenal (HPA) axis negative feedback loop was disrupted in chronic stress, thus resulting in elevated cortisol level27. It implies that there was absent of diurnal variation. Another study done among white collar workers to determine the relationship between long term job stress and morning and evening salivary cortisol revealed that there was a sharp drop of morning salivary from weekdays to weekends. A possible explanation for the failed findings may be that increased morning cortisol reactivity was influenced more by acute rather than by chronic stress exposure 28. Hence, a relationship between acute stress and salivary cortisol level may give a more valid result. |
Limitations
|
| The main limitation was the cross-sectional design of this study. The validity of study results was compromised by factors such as the single time point measurements and difficulties in ascertaining the causal relationship. Studies permitting temporal sequence such as cohort or case-control studies would be useful. However, it would be beyond the resources available for this study. Secondly, the validity of the data related to the variables cannot be determined because the information obtained was solely through selfreported measure. As defined in the literature, self-reported data cannot be viewed as objective29. Thirdly, the sample consisted of a specific group of employees, and therefore the generalizability of the results beyond the study population should be cautiously done. However, the usefulness of recruiting specific groups may expand our understanding of the stress process for that particular group of workers. Hence, the organization would be able to make use of the results for further action in prevention of stress among their workers. |
Recommendation
|
| Joint workplace initiatives should be taken by both employees and employers (the organization) in accordance to International Labour Organization (ILO) convention concerning safety and health and working environment30. Designing jobs to give assistant medical officers more decisionmaking authority and skill discretion are indicated to reduce stressor, hence decrease their stress. Practising team-oriented approaches will also contribute to employee well-being among health care workers. Further study to determine relationship between acute stress and salivary cortisol level may give a more valid explanation regarding these relationships. |
Conclusions
|
| The prevalence of stress among assistant medical officer was 13.7%. Salivary cortisol, the biomarkers of stress was significantly higher among stressed compared to nonstressed assistant medical officers. However, there was no correlation between salivary cortisol and stress score which was probably due to the low prevalence of stress among assistant medical officers in Ministry of Health hospitals in Kelantan and Terengganu. |
Competing Interest
|
| None of the authors has any conflict of interest by publishing this article. |
Authors’ Contribution
|
| All the authors have contributed significantly in the supervision, analysis, drafting and revising the article to its completion. They take full responsibility for its content and have read the manuscript fully. |
Acknowledgements
|
| We thank the Ministry of Health, Malaysia, Kelantan and Terengganu State Health Department and the assistant medical officers for their cooperation. |
Ethical Policy
|
| The research protocol was approved by the Research and Ethical Committee Universiti Sains Malaysia [USMKK/PPP/JePeM[202.4.(2.5)]] and Ministry of Health Malaysia [bil.(38)dlm.JKN.TR.7/23/2Pt.6] and informed consent was obtained from each subject prior to inclusion in the study. |
 |
 |
 |
 |
References
|
- Marine A, Ruotsalainen JH, Serra C,Verbeek JH. Preventing occupational stressin healthcare workers review). TheCochrane Collaboration 2009.
- Weinberg A, Creed F. Stress andpsychiatric disorder in healthcareprofessionals and hospital staff. The Lancet.2000;355:533-537.
- Sutherland VJ, Cooper CL. Job stress,satisfaction and mental health among generalpractitioners before and after introduction ofnew contract. British Medical Journal.1992;13: 1545-1548.
- Tham KY, Tan YH, Loh OH, Tan WL,Ong MK, Tang HK. Psychiatric morbidityamong emergency department doctors andnurses after the SARS outbreak. Ann AcadMed Singapore 2004; 33: S78-9.
- Okada N, Ishii N, Nakata M,Nakayama S. Occupational stress amongJapanese emergency medical technicians:Hyogo prefecture. Prehospital Disaster Med. 2005; 20:11-121.
- Kirschbaum C, Prussner JC, Stone A,et al. Persistent high cortisol responses to repeated psychological stress in asubpopulation of healthy men.Psychosomatic Medicine. 1995;57:468-474.
- De Schippera EJ, Marianne JRW,Geurtsb SAE, de Weertha C. Cortisol levelsof caregivers in child care centers as relatedto the quality of their caregiving. EarlyChildhood Research Quarterly. 2009;24.
- Koh D, Lim MK, Chia SE, et al. Riskperception and impact of Severe AcuteRespiratory Syndrome (SARS) on work andpersonal lives of healthcare workers inSingapore: what can we learn? Med Care.2005;43:676-682.
- Koh DSQ, Koh GCH. The use ofsalivary biomarkers in occupational andenvironmental medicine. Occup Environ Med 2007;64:202-210.
- De Weerth C, Graat G, Buitelaar JK,Thijssen JHH. Measurement of cortisol insmall quantities of saliva. Journal of Clinical Chemistry. 2003;49:658-659.
- Daniel WW. Determination of samplesize for estimating proportion. In:Biostatistics: a foundation for analysis in theHealth Science. 7th edition. 1999:183-184.
- Azlihanis AH. Prevalence andassociated factors of depression, anxiety andstress among secondary school teachers inKota Bharu, Kelantan (dissertation).UniversitiSains Malaysia, KubangKerian.2007.
- Lovibond SH, Lovibond PF. Manualfor the depression anxiety stress scales,Psychology Foundation Monograph, Sydney.2002.
- Miller RL, Pallant JF, Negri LM.Anxiety and stress in the postpartum: is theremore to postnatal distress than depression?BMC Psychiatry. 2006;6. Available at:https://www.biomedcentral.com/1471-244X/6/12.
- Salimetrics. Expanded range highsensitivity salivary cortisol enzymeimmunoassay kit. 2006. Available at:www.salimetric.com.
- Harmy MY. A comparative study ofjob stress among nurses in HospitalUniversity Science of Malaysia and femaleclerical workers in Kota Bahru districtKelantan (dissertation). UniversitiSains
- Malaysia, KubangKerian. 2001.
- Michael G, Anastosios S, Helen K,Catherine K, Christine K. Gender differencesinexperiencing occupational stress: the roleof age, educaton and marital status. Stressand Health. 2009. Available at:https://www.interscience.wiley.com.
- Aziah BD, Rusli BN, Winn T, NaingL, Tengku MA. Prevalence and associatedfactors of job-related depression inlaboratory technicians in Hospital UniversitiSains Malaysia (HUSM) and Ministry ofHealth Malaysia (MOH) Hospitals inKelantan. Malaysia Medical Journal.2004;59:268-278.
- Shen HC, Cheng Y, Tsai PJ, Lee SHS,Guo YL. Occupational stress in nurses inpsychiatric institutions in Taiwan. Journal ofOccupational Health. 2005;47:218-225.
- Shimizu T, Hiro M, Mishima N,Nagata S. Job stress among Japanese fulltimeOccupational Physicians. Journal OfOccupational Health. 2002:348-354.
- Weekers NY, MacLean J, Berger DE.Sex, stress and health: does stress predicthealth symptoms differently for two sexes?Stress and Health. 2005;21:147-156.
- Rusli BN, Edimansyah BA, Naing L.Prevalence and associated factors of stress indental healthcare workers of a higherinstitution of learning in Kelantan. Archivesof Orofacial Sciences. 2006;1:51-56.
- Yang Y, Koh D, Ng V, et al. Salivarycortisol levels and work-related stress amongemergency department nurses. Journal ofOccupational and Environmental Medicine.2001;43: 1011-1018.
- Vedhara K, Miles J, Bennett P, et al.An investigation into the relationshipbetween salivary cortisol, stress, anxiety anddepression. Biological Psychology.2003;62:89-96.
- Steptoe A, Ussher M. Smoking,cortisol and nicotine. International Journal ofPsychophysiology. 2006;59:228-235.
- Chaudhuri SK. Concise medicalphysiology: New Central Book Agency.1991.
- Schultze EA, Laudenslager ML,Coussons-Read ME. An investigation intorelationship between stress, depression anddiurnal salivary cortisol in Western Coloradoranchers. Brain, Behavior, and Immunity.2005;19:e63.
- Rydstedt LW, Cropley ML, DevereuxJJ, Michalianou G. The relationship betweenlong-term job strain and morning andevening saliva corisol secretion amongwhite-collar workers. Journal ofOccupational Health Psychology.2008;13:105-113.
- Podsakoff PM, Organ DW. Self-reportin organization research: problems andprospects. Journal of Management.1986;12:531-544.
- Smith J, Albracht G, Coninck SD.Labour and the environment: a naturalsynergy. United Nations EnvironmentalProgramme, WHO. 2007;1-152.
|
