Prevalence of Mental Disorders among Shift Work Hospital Nurses in Shiraz, Iran

MATERIALS AND METHODS

Study location and subjects: In this study, names and addresses of all governmental and non-governmental hospitals of Shiraz were prepared. All hospitals were visited. The objectives of the study were explained for the managements of the hospitals and the head nurses to obtain their agreement. In each hospital, a list of all registered nurses was prepared by the head nurses and given to the researchers. Then, copies of a questionnaire as the collecting data tool were distributed among the nurses in the corresponding list.

The nurses were asked to fill out the questionnaire if they agreed to participate. From 1396 registered nurses employed in all 12 hospitals in Shiraz, 1195 nurses including 1064 female and 131 male partook in this study. Since high percent of the registered nurses participated (85.6%), the results of this study could be generalized to all nurses of Shiraz hospitals. Data were collected from September to October, 2006.

Data collection: The 28-item General Health Questionnaire (GHQ-28) was used as a self administered screening tool for detection of mental disorders. This questionnaire was developed by Goldberg and Hillier (1979) for screening somatic symptoms, anxiety and insomnia, social dysfunction and severe depression. It has been widely used in many countries in clinical and nonpsychosis mental disorders (Adachi et al., 1999; Fagin et al., 1995; Gage and Leidy, 1991; Holland, 1996; Hunter and Houghton, 1993; Schreuders et al., 2007; Spelten et al., 1993). Studies on the validation of the GHQ-28 in different countries have demonstrated its high validity and reliability as a screening tool for mental disorders in the community (Ross et al., 1988; Lykouras et al., 1996; Failde et al., 2000).

In this study, Persian version of General Health Questionnaire (GHQ-28) was used. Noorbala et al. (1999) reported that the sensitivity specificity and overall misclassification rate of the Persian version of GHQ-28 were 84.7, 93.8 and 82%, respectively, for a cut-off point of 6. Each item was scored based on a 4-point scale (i.e., always to never). Regarding to the interclass correlation between the test-retest scores (0.85), for each subscale cut-off point 6 and for total score cut-off point 23 were applied in the present study (Moayedi, 2005). By this means, those scoring 6 or above were designated as possible case of mental disorder. This survey was approved by the Ethics Committee of Tehran University of Medical Science, prior to its commencement.

In each hospital, the Persian version self-administrated GHQ-28 was distributed among all nurses and collected by the nursing center director. The participants were assured that the personal information would be treated with complete confidentiality.

Statistical methods: Upon completion of field survey and data collection, data was coded and transferred into the computer for further analysis. Statistical analyses were performed using SPSS version 11.5.

Two sample t-tests were used to determine the factors that affect mental disorders. Mental disorders were considered as dependent variables and gender, education, marital status, shift work and type of hospital were considered as independent variables.

RESULTS AND DISCUSSION

Demographic variables: The mean (SD) age of nurses was 30.81 (7.08) years; 89.04% of the participates was female and 10.96% was male. One thousand and seventy three of the nurses studied (89.79%) worked on rotating shift system and 122 nurses (10.21%) worked on fixed shift (day time). The mean (SD) job tenure of the nurses in the current hospital was found to be 7.28 (SD = 6.83) years. 50.46% of the participants were married and 62.85% worked in non-governmental hospitals (Table 1).

GHQ-28 Questionnaire results: The results of the GHQ-28 Questionnaire indicated that 45.4% of the nurses studied suffered from some form of mental disorders. It was more common among females than males. As shown in Table 2, the percentage of those who scored between

Table 1:	Some personal details of the nurses who participated in the study (n = 1195)

Table 2:	Distribution of GHQ-28 scores (n = 1195)

0-22 points (considered to be mentally in good health) was 54.6%; the remaining 45.4% scored 23 to 84 points (considered to be mentally in poor health). As indicated in Table 3, nurses scored highest on the social dysfunction subscale (mean = 8.89, SD = 3.42), followed by the anxiety symptoms sub scale (mean = 6.44, SD = 4.48).

Mean and standard deviation of subscales (including, depression, anxiety, somatisation, social dysfunction and mental disorders) based on gender, marital status; shift working and type of hospital are depicted in Table 4. As shown in this Table 4:

Table 3:	Mean and standard deviation of subscales score as studied mental health status of the hospital nurses (n = 1195)
AStandard deviation

The results also showed that there was no significant difference between mental disorders and job tenure (p = 0.413).

Few occupational health studies have been conducted on the prevalence of the mental disorders among hospital nurses in Iran. This survey was implemented to study more thoroughly on nurses` subjective mood and mental health status in Shiraz and therefore represented the first such large-scale study. The results revealed that 45.4% of nurses studied were at risk of developing mental disorders. It is to be noted that the nurses experienced depression and anxiety disorders commonly associated with work organization and work demands, such as shift working, meeting the demands of supervisors and physicians, dealing with the conflicting demands of administration and management and meeting the needs of patients.

Comparing the results of local studies on other target groups such as adult population or female workers with the results of this study showed that the prevalence rates of mental disorders among nurses studied were higher than the rates in Noorbala et al. (2004) and Ahmad-Nia (2002) studies (Table 5). This indicates that nursing job can be considered as a high risk occupation for developing mental disorders.

Social dysfunction was the most prevalent disorder among nurses studied (79.5%) which was significantly

Table 4:	Mean and standard deviation of subscales and scores in gender, marital statues, shift working and type of hospital groups (n = 1195)
ADepression; BAnxity symptous; CSomatisation; DSocial dysfunction; EMental disorders *Independent t-test

Table 5:	Comparison of prevalence of depression, anxiety symptoms, somatisation, social dysfunction and mental disorders in general Iranian population, Ahmad-Nia study (2002) and in the study population
ATest of proportion (difference between study population and general Iranian population) BTest of proportion (difference between study population and Ahmad-Nia (2002) study population)

higher than its rate in general Iranian population with prevalence rate of 14.2% (p<0.05) (Noorbala et al., 1999). This could be attributable to high workload leading to fatigue and exhaustion (Rashidinik, 2001).

In an epidemiological study on Iranian population (2001), the prevalence rate of anxiety disorders and depressive symptoms were reported to be 8.48% and 2.26%, respectively (Ahmad-Nia, 2002), while in the present study those rates were found to be 43.2 and 11.2%, respectively, which were significantly higher (p<0.0001).

The prevalence rates of mental disorders obtained in this study were lower than rates reported by Suzuki et al. (2004), but higher than those obtained by Mori and Kageyama (1995).

The present study also found that female nurses were 2.7 times more at risk of mental disorders as compared with male nurses. These findings were consistent with the results of other studies (Mori and Kageyama, 1995; Suzuki et al., 2004; Kessler, 1994; Fakhari et al., 1999). Association was also observed between sex and the subscales including, anxiety and depression. The mentioned problems were more prevalent among male than female. This is in accord with the result of Kessler (1994).

No significant association was found between mental disorders and age and job tenure. Social dysfunction was significantly associated with age such that social dysfunction increased with age increasing. Stansfeld and Marmot (1992) believed that mental disorders occurred

less, particularly in females, when job tenure was low.

The results of the present study also showed that depression was more prevalent among single nurses, but social dysfunction occurred more prevalent among married nurses (p<0.05). Fakhari et al. (1999) pointed out that the rate of mental disorders among married was 1.59 times higher than that of singles. The study revealed a significant association between shift work and anxiety (p<0.05). These findings were in agreement with the results of other studies such as Suzuki et al. (2004) and Mori and Kageyama (1995).

CONCLUSION

In conclusion, the results of the present study showed that mental health status of the hospital nurses was poor, as shown by mean GHQ-28 score. This study also supported the results of previous studies showing high rates of mental disorders among hospital nurses.

The mental health pattern among hospital nurses is similar to the national information on the prevalence of mental disorders, but it seems that the prevalence of social dysfunction and anxiety disorders in the nurses is higher than Iranian population. Paying more attention to shift work nurses` health, particularly female nurses, through shift work health education program, occupational health consultation and periodical examinations is recommended.

ACKNOWLEDGMENTS

Funding through Tehran University of Medical Sciences via contract No. 132/6431 supported this study. The authors wish to thank managers, supervisors and all nursing staff in Shiraz hospitals for their valuable co-operation.