Gender differences in the psychological impact of tsunami

Abstract

Aim:

The aim of this observational study was to explore gender-related differences in psychiatric morbidity during the initial three months following the December 2004 earthquake and tsunami involving the Andaman and Nicobar Islands, India.

Methods:

There were 12,784 survivors sheltered across 74 relief camps with 4,684 displaced survivors in Port Blair and 8,100 non-displaced survivors in Car-Nicobar Island. All persons who accessed mental health assistance within the camps constituted the study sample. Diagnoses were made by qualified psychiatrists using the ICD-10. There were 475 patients: 188 (40%) men and 287 (60%) women.

Results:

There were significant gender differences in terms of displacement. There were significantly higher levels of panic disorder, unspecified anxiety disorder and somatic complaints in the displaced women while the non-displaced population showed more adjustment disorder.

Conclusions:

Displacement was a significant factor in the manifestations of observed pathology. Displaced women had greater psychiatric morbidity. In addition, the fact that adjustment disorder (a self-limiting disorder form of psychopathology) was more prevalent in the non-displaced group may be a reflection of the findings of overall lesser morbidity in non-displaced women. Hence, women may have to be rehabilitated in their own habitats after major disasters.

Keywords

Disaster gender tsunami high-risk population mental health

Introduction

Natural disasters can have an adverse impact on the psychosocial well-being of persons exposed to them. Disasters pose a threat to personal safety, stress out defence mechanisms and cause disruption of community and family structures (Lubit & Spencer, 2003). These events may also cause mass destruction of life and property with the resultant loss of social networks and daily routines of the individuals involved. The majority of people affected by a disaster show a typical pattern of mental, emotional and physical response (Carol, 2003; World Health Organization, 1992a). The symptoms severity mainly depends on individual and sociocultural factors, and the extent of the disaster (Brewin, Andrews, & Valentine, 2000; Carlier & Gerson, 1997). Individual factors include coping patterns, pre-morbid personality and other individual resilience factors. Sociocultural factors include existing social and family support, vulnerability of the special populations, and community characteristics (Hutton & Haque, 2004; Johnes, 2000; Morrow, 1999; North, Smith, & McCool, 1989; Regehr, Hemsworth, & Hill, 2001). Increased risk of adverse outcome in disaster survivors is most consistently associated with: (1) severity of the exposure to the disaster; (2) community being highly disrupted or traumatized; (3) lack of adequate social support; (4) being an ethnic minority; (5) post-traumatic stress; (6) female gender; and (7) extremities of age (children and elderly) (Brewin et al., 2000; Norris, Friedman, & Watson, 2002; Palinkas, Downs, Petterson, & Russell, 1993).

No disaster is the same as another; similarly, each disaster survivor has unique experiences and different people in the same community respond differently (Almedon & Summerfield, 2004; Summerfield, 1999). Ethnic and racial minority groups are likely to have an increased risk of adverse outcomes. The reasons for this may be language barriers, rejection of outside interference, non-availability of assistance (especially with regard to mental health), displacement, differences in socioeconomic status, religion, spiritual practices and cultural values, all of which pose barriers for access to treatment and psychosocial rehabilitation (Jaycox, Marshall, & Schell, 2004; Norris & Algeria, 2005; Perilla, Norris, & Lavizzo, 2002; Snowden, 2003).

India is a land known for its diversity; people differ in religion, language, culture and ethnicity. These diversities may influence the need for and availability of help, comfort in seeking help and acceptability of the available help (Carballo, Heal, & Hernandez, 2005; Norris & Algeria, 2005) Hence, any planned intervention should be appropriate to the local needs (Barron, 2004). In addition, agencies from outside, foreign relief efforts and the media can worsen the existing status of individuals and their communities (Hawkins, McIntosh, & Silver, 2005). Lack of trained manpower in the mental health field and widespread stigma and discrimination add to the difficulty in the provision of mental health care (Chandrashekar & Math, 2006). In the face of limited resources, it is important to identify people at risk for mental health problems to allocate resources for their care (Somasundaram & van de Put, 2006).

Women are more vulnerable than men to the traumatic events of disasters (Aksaray, Kortan, & Erkaya, 2006; Heir et al., 2010; Murphy, 2010). Post-disaster psychological reactions like major depression, post-traumatic stress disorder (PTSD), anxiety and somatization have been found to more prevalent in women than men (Aksaray et al., 2006). In addition, women are also exposed to gender-based violence during disasters, which is associated with poorer mental health outcomes (Anastario, Shehab, & Lawry 2009a, 2009b; Rosborough, Chan, & Parmar, 2009). Studies from developing countries regarding female survivors are scarce. Hence, this study explored the gender-related differences in psychiatric morbidity during the initial three months following the tsunami involving the Andaman and Nicobar Islands, India.

Context of the study

An earthquake measuring 9.0 on the Richter scale triggered a massive tsunami in the early hours (06:00 hours, Indian standard time) of 26 December 2004, and devastated the population of the Andaman and Nicobar Islands of India in the Bay of Bengal. The population of the Andaman and Nicobar Islands is comprised of the tribal population native to the islands and settlers from various states of the Indian mainland. While the destruction of homes and property led to the relocation of a large number of tribals and settlers to relief camps in Port Blair, a significant section of the indigenous tribal population and a smaller number of settlers chose to stay on their home islands.

The National Institute of Mental Health and Neuro Sciences (NIMHANS), in Bangalore, India, was designated by the government of India as the nodal agency for the assessment and coordination of psychosocial relief to the affected population in India following the tsunami. Hence, a multidisciplinary team was deputed to the Andaman and Nicobar Islands to, mainly, identify and treat the persons in the camp needing immediate intervention, and assess psychiatric morbidity immediately after the event. The data presented in this study is a part of this community assessment and intervention during the relief work.

Methods

Subjects

The study was undertaken in the 12,784 survivors sheltered across 74 relief camps in the area. All those persons who accessed mental health assistance within the camp constituted the study sample. Port Blair had 12 camps that provided shelter to approximately 4,684 displaced survivors (displaced survivors group). There were 62 camps on Car-Nicobar Island that provided shelter to approximately 8,100 survivors. These survivors continued to stay in their own island and were not displaced from their habitats (non-displaced survivors group).

Screening

The team visited the 74 relief camps that sheltered 12,784 survivors and screened and assessed them for mental health problems. The affected population was large, and the survivor population in each camp was highly mobile and heterogeneous. Considering the logistics, the team approached each camp and contacted the medical officer in charge, the survivors’ community leaders and the local staff of the camp. They were educated about the mental health consequences associated with a disaster and the need and process for establishing mental health clinics for the survivors inside the camp was discussed. The consent and cooperation of the community leaders was obtained, as this was essential for community participation and an entry point into the community. Following this, several announcements were made by the community leaders and the local staff of the camp about the availability of mental health rofessionals, along with descriptions of common signs of pathology for which persons were encouraged to seek help. Subsequently, mental health clinics were established inside the relief camps. During the consultations with the survivors, attempts were also made to identify other cases using the snowball technique. The snowball technique is a special non-probability method that relies on referrals from initial subjects to generate additional subjects. As a part of assessment and intervention, the team adopted a multi-tiered approach encompassing disaster mental health awareness programmes inside the camp, group discussions, poster displays, the training of primary health care physicians, nurses and workers in basic diagnostic skills and counselling, and the education of key community members such as captains, teachers, religious leaders, relief workers, social workers and volunteers who were survivors.

Diagnoses

A qualified psychiatrist diagnosed the patients using the ICD-10 criteria (World Health Organization, 1992b). If there was any doubt regarding the diagnosis, it was reviewed by a senior psychiatrist and a consensus was sought. Data were collected using a semi-structured, clinical proforma that was developed and had been in use at the screening and outpatient clinic at NIMHANS for decades. This proforma was used to capture basic demographic details, history of presenting illness, family history, personal history and mental status. Consultations usually were performed in the open air inside the camp. Informed oral consent was obtained from all the subjects. The study was conducted in accordance with the guidelines provided by the NIMHANS Ethics Committee.

Statistical analyses

The statistics were processed using SPSS Version 11 (SPSS Inc., Chicago). The socio-demographic details were assessed using descriptive measures and compared between groups. The χ ² and Fisher exact tests were used to process the data. All results with p < .05 were considered to be statistically significant.

Results

Socio-demographic details have been presented as descriptive frequency tables comparing men and women (Table 1). Of the total number of subjects (N = 475), there were 188 men (40%) and 287 women (60%). The age distribution in men (M ± SD = 40.6 ± 18.2) and women (M ± SD = 40.7 ± 15.8) was similar (t = -0.07, p = .9). The men were more often displaced (68%), while women were mostly non-displaced (60%). This difference was statistically significant (p < .001).

Table 1

Sociodemographic profile.

		Gender n (%)
		Male 188 (40)	Female 287 (60)
Religion	Hindu	107 (57)	96 (33)
	Christian	71 (38)	182 (63)
	Muslim	10 (5)	9 (4)
Displaced vs undisplaced	Displaced	128 (68)	116 (40)
	Undisplaced	60 (32)	171 (60)
Indigenous vs settler population	Indigenous population	73 (39)	176 (61)
	Settler population	115 (61)	111 (39)

A comparison of men and women did not show any significant difference in terms of clinical diagnoses (Table 2). Considering the fact that there was a significant difference between the two groups in terms of displacement, and as our previous study showed that displacement was one of the variables that predicted morbidity, we decided to compare the two groups in the displaced versus non-displaced populations separately. The comparisons included displaced men versus displaced women, non-displaced men versus non-displaced women, displaced men versus non-displaced men and displaced women versus non-displaced women. Significant differences were found only in the comparison between displaced versus non-displaced women (Table 3). Within the female population, there were significantly higher levels of panic disorder (p = .001), unspecified anxiety disorder (p = .004) and somatic complaints (p = .004) in the displaced group (p = .001), while the non-displaced population showed more adjustment disorder (p = .001). However, these differences could also be a function of other factors such as religion, residence or indigenous versus settler population variations. Hence, binary logistic regression analyses were done with aforementioned psychiatric diagnoses (panic disorder, adjustment disorder, unspecified anxiety disorder and somatic complaints) as dependent variables and displacement, religion, indigenous versus settler population and residence as independent variables. Displacement emerged as the single most important factor for panic disorder (β = 1.33, SE = 0.40, p = .001, OR = 3.77) and adjustment disorder (β = −1.28, SE = 0.27, p < .001, OR = 0.28), whereas being a settler was the significant predictor for unspecified anxiety disorder (β = 2.15, SE = 0.79, = .006, OR = 8.62) and somatic complaints (β = 2.61, SE = 1.07, p = .02, OR = 13.59).

Table 2.

Gender differences in clinical characteristics.

Diagnoses	Gender n (%)		p (Fischer’s I)
Diagnoses	Male 188 (40)	Female 287 (60)	p (Fischer’s I)
Adjustment disorder	62 (33)	116 (40)	0.12
Acute stress reaction	2 (1)	1 (0.3)	0.57
PTSD	24 (13)	29 (10)	0.38
Depression	32 (17)	70 (24)	0.07
Panic disorder	25 (13)	32 (11)	0.48
Somatic complaints	6 (3)	9 (3)	1.00
Phobic disorder	3 (2)	1 (0.3)	0.31
Anxiety disorder unspecified	14 (7)	12 (4)	0.15
Schizophrenia	3 (2)	3 (1)	0.69
Dissociative disorder	–	2 (0.7)	0.52
Acute psychosis	1 (0.5)	–	0.40
Dysthymia	–	3 (1)	0.28
Recurrent depressive disorder	0	2	0.52
No ICD-10 diagnosis	9	7	0.20

Table 3.

Comparison of displaced vs non-displaced women (n = 287).

Diagnoses	Displacement n (%)		P (Fischer’s t)
	Displaced 116 (40)	Non-displaced 171 (60)
Adjustment disorder	27 (23)	89 (52)	< 0.001***
Acute stress reaction	–	1 (0.6)	1.00
PTSD	14 (12)	15 (9)	0.43
Depression	28 (24)	42 (25)	1.00
Panic disorder	22 (20)	10 (6)	0.001**
Somatic complaints	8 (7)	1 (0.6)	0.004**
Phobic disorder nos	–	1 (0.6)	1.00
Anxiety disorder nos	10 (9)	2 (1)	0.004**
Schizophrenia	–	3 (2)	0.28
Dissociative disorder	–	2 (1)	0.52
Dysthymia	–	3 (2)	0.28
Recurrent depressive disorder	0	2	0.52
No ICD-10 diagnosis	7	0	0.002**

p < 0.05, ** p < 0.01, *** p < 0.001

Discussion

It is to be kept in mind that the information, being derived from records of clinical interventions carried out during the early phase of the disaster, has the limitation that assessments using structured instruments and rigorous sampling methods could not be carried out. The nature of the study is observational, which overrules the measure of actual prevalence of disaster-related disorders. Only patterns of morbidity have been identified. However, the main strength of the results is that both the community leaders and the survivors were involved in the shared decision-making processes with regard to organizing and utilizing the services at the mental health clinics inside the relief camps. The diagnosis was made by a qualified psychiatrist using the ICD-10 criteria.

The fact that there was no significant gender difference in patterns of psychiatric morbidity is in contrast to other studies that have found a greater occurrence of major depression, PTSD, anxiety and somatization in females post-disaster (Aksaray et al., 2006; Heir et al., 2010; Murphy, 2010). However, on further analysis, it was found that displacement was a significant factor in the manifestations of observed pathology. A majority of women preferred non-displacement (60%) in favour of displacement. In addition, most women who displaced were settlers (91%). Indigenous women remained in their own habitats and were mostly respected, protected and given immediate priority in accessing resources. It was found that displaced women had significantly higher psychiatric morbidity in the form of panic disorder, unspecified anxiety disorder, somatic complaints and others. However, the higher occurrence of unspecified anxiety disorder and somatic complaints were associated with the settler status of the displaced women rather than displacement per se. As previously stated, the settler women were given lower priority in relation to indigenous women in terms of access to resources. The settler women were expecting money in terms of rehabilitation, whereas the indigenous population did not expect anything from rehabilitation. Cultural and resilient factors also would have played a crucial role in indigenous women; they were all well supported in a joint family system, by altruistic community leaders and religious/spiritual groups, and less dependent on contemporary materialistic world needs.

In contrast, adjustment disorder was more common in the non-displaced women. The fact that adjustment disorder (a lesser form of psychopathology) was more prevalent in the non-displaced group may be a reflection of the findings of overall lesser morbidity in non-displaced women. However, depression and PTSD were distributed equally in both groups. These findings are somewhat similar to the study done in Thailand (van Griensven, Chakkraband, & Thienkrua, 2006) and may indicate that women need to be rehabilitated in their own habitats after major disasters.

Inherent to the state of displacement away from neighbours and familiar environments is the loss of social support due to the loss of community. This is likely to facilitate greater feelings of isolation and a state of anomie. This classic sociological concept describes a state of normlessness and loosening of social norms, which is manifested in such social maladies as suicide and riots in what were previously socially and civilly cohesive and functioning environments (Giddens, 1972). Related to anomie is a reduced sense of control over one’s own life. A sense of community – including residing within known social groups, familiar cultural activities, structured social relationships and established roles – mitigates against a sense of normlessness (Putnam, 2000). These elements are even more important in women.

In addition to the above findings, there were many other gender-relevant issues noted by the relief team members during the interaction with women survivors within the relief camp. There was a lack of privacy for women inside the relief camps that caused them enormous stress. Rehabilitation relief workers were not sensitive about the needs of women. Many of the women raised concerns about the inadequate privacy for bathing, toilet and changing rooms. They also spoke about the discrimination and difficulty competing with men in access to resources, which was strikingly noted inside the relief camp.

Interaction with women survivors revealed that the pre-disaster responsibilities of women within the four walls were magnified by the onset of disaster. Women in developing countries face inequality in all spheres in day-to-day life. This was exaggerated during the disaster situation. Women were the primary caretakers of the other family survivors such as children and the elderly. The lack of support and resources compounded the pre-disaster traditional roles and familial responsibilities of the local woman.

Women expressed helplessness and were more vulnerable for exploitation by the loss of men, especially when a male head of household had died or was missing. This has been noted by many previous studies (Anastario et al., 2009b; Rashid & Michaud, 2000). Women had to live in constant fear of sexual abuse, abduction, rape and violence. They felt extremely vulnerable because of the chaotic environment following the disaster where they did not know whom to call for help or whom they could trust. Even the disaster relief team was initially viewed with suspicion and mistrust. Pregnant women, breast-feeding mothers, widows, the physically and mentally disabled and the elderly made up particularly vulnerable groups in the disaster situation. Many of the women were anaemic and malnourished, and deterioration in their health had a direct impact on the health of their children. To compound their misery, many male survivors utilized all the monetary relief in consuming alcohol and other substances on the pretext of coping with disaster loss. Unfortunately, this substance use was also closely associated with violence against women. Poor financial condition, insufficient social and family support and an atmosphere of total insecurity bred psychopathology in the female population.

Non-availability of female doctors in the disaster relief team played a crucial role in the help-seeking behaviour of women survivors. The majority of the post-disaster relief work was male-centred, such as providing money, job, livelihood resources and so forth. Women were mostly kept outside the purview of accessing these rehabilitation resources. A lack of concern and sensitivity about the various needs of women, such as providing sanitary towels and childcare materials, such as nappies, feeding bottles, mosquito nets and so forth, was noted as a barrier to relief management. Few female relief workers were available to provide care, which added to the challenges.

There is a need to sensitize relief and health workers about women’s health needs, post-disaster. Many suggestions have been proposed to address the potential needs of women, such as involving women in disaster planning, the provision of gynaecological and obstetric supplies, and encouraging local women to participate in disaster relief work (Krajeski & Peterson, 1999; Meyers, 1994; Richter & Flowers, 2008). The axiom of ‘women and children first’ has to be implemented when dealing with major disasters (Meyers, 1994). The inclusion of women’s specialists in relief work will reduce gender bias and increase the effectiveness of disaster management efforts.

Early warning systems are poor in developing countries. Even if the early warning systems work, they fail to reach women and children. Women are less informed about the disaster events and the location of relief materials, and they are also not involved in the disaster management. Hence, it is essential to involve women in the training, from disaster preparedness to disaster management. They also should be empowered about how to deal with and protect children in a disaster crisis. Also, the relief supplies delivered do not always address the needs of women who take care of the family, which includes children, the elderly and the sick. Hence, it is essential to involve women in needs assessment and decision-making process and also in providing relief. During rehabilitation and when providing relief work, women and children should be the first priority. Women need to be involved and empowered in the process of community rebuilding and rehabilitation. Educating and empowering women equals educating a family. They are an asset in providing care to children, the injured and the elderly. Hence, neglecting and not involving women in the disaster management means not addressing and providing care to the most vulnerable and needy. As women and children are more vulnerable and exposed to sexual and domestic violence during a disaster situation, adequate protection needs to be in place to protect them from predators, especially in the relief camps.

Conclusion

In conclusion, the medical and mental health needs of women are crucial and need to be addressed to decrease post-disaster morbidity. Women may need to be rehabilitated in their own habitats after major disasters. Sensitization of the relief workers and administration, as well as involvement of women in disaster planning and disaster relief work, is essential to reduce gender inequality and provide better care.

Footnotes

Acknowledgements

We would like to thank the Departments of Health, Education and Social Welfare of the Andaman and Nicobar Islands, India. We also would like express our gratitude to all the relief team members from various organizations such as the All India Institute of Medical Sciences, Delhi, and the nursing staff from various hospitals, non-governmental organizations and other agencies for providing relief during the tsunami.

This paper was presented as a poster in the Brain, Behaviour and Mind – Advancing Psychiatric Care in the East, Second HK-UK International Conference: Moving on – From Science to Service

Conflict of interest

Part of the data have been published previously in Math et al. (2006) and . The data presented in this manuscript have been re-analysed from a gender perspective, which is not published elsewhere.

References

Aksaray

Kortan

Erkaya

(2006). Gender differences in psychological effect of the August 1999 earthquake in Turkey. Nordic Journal of Psychiatry, 60, 387–391.

Almedon

A. M.

Summerfield

(2004). Mental well-being in settings of ‘complex emergency’: An overview. Journal of Biosocial Science, 36, 381–388.

Anastario

Shehab

Lawry

(2009a). Increased gender-based violence among women internally displaced in Mississippi 2 years post-Hurricane Katrina. Disaster Medicine and Public Health Preparedness, 3, 18–26.

Anastario

Shehab

Lawry

(2009b). Responding to gender-based violence in disasters. Disaster Medicine and Public Health Preparedness, 3, 138–139.

Barron

R. A.

(2004). International disaster mental health. Psychiatric Clinics of North America, 27, 505–519.

Brewin

C. R.

Andrews

Valentine

J. D.

(2000). Meta-analysis of risk factors for posttraumatic stress disorder in trauma exposed adults. Journal of Consulting and Clinical Psychology, 68, 748–766.

Carballo

Heal

Hernandez

(2005). Psychosocial aspects of the Tsunami. Journal of Royal Society of Medicine, 98, 396–399.

Carlier

Gerson

(1997). Stress reactions in disaster victims following the Bijlmermeer plane crash. Journal of Traumatic Stress, 10, 329–335.

Carol

S. N.

(2003). Psychiatric epidemiology of disaster responses. In Ursano

R. J.

Norwood

A. E.

(Eds), Trauma and disaster, responses and management: Review of psychiatry series, volume 22 (pp. 35–62). New York: American Psychiatric Publishing.

10.

Chandrashekar

C. R.

Math

S. B.

(2006). Psychosomatic disorders in developing countries: Current issues and future challenges. Current Opinion in Psychiatry, 19, 201–206.

11.

Giddens

(1972). Emile Durkheim: Selected writings. London, UK: Cambridge University Press.

12.

Hawkins

N. A.

McIntosh

D. N.

Silver

R. C.

(2005). Early responses to school violence: A qualitative analysis of students’ and parents’ immediate reactions to the shootings at Columbine High School. Journal of Emotional Abuse, 4, 197–223.

13.

Heir

Rosendal

Bergh-Johannesson

Michel

P. O.

Mortensen

E. L.

Weisaeth

…Hultman

C. M.

(2010). Tsunami-affected Scandinavian tourists: Disaster exposure and post-traumatic stress symptoms. Nordic Journal of Psychiatry, 65, 9–15.

14.

Hutton

Haque

C. E.

(2004). Human vulnerability, dislocation and resettlement: Adaptation process of riverbank erosion-induced displacees in Bangladesh. Disaster, 28, 41–62.

15.

Jaycox

L. H.

Marshall

G. N.

Schell

(2004). Use of mental health services by men injured through community violence. Psychiatric Services, 55, 415–420.

16.

Johnes

(2000). Aberfan and the management of trauma. Disasters, 24, 1–17.

17.

Krajeski

R. L.

Peterson

K. J.

(1999). ‘But she is a woman and this is a man’s job’: Lessons for participatory research and participatory recovery. International Journal of Mass Emergency Disasters, 17, 123–130.

18.

Lubit

Spencer

(2003). Children, disasters, and the September 11th World Trade Center attack. In Ursano

R. J.

Norwood

A. E.

(Eds), Trauma and disaster, responses and management (pp. 63–96). New York: American Psychiatric Publishing.

19.

Math

S. B.

Girimaji

S. C.

Benegal

Uday Kumar

G. S.

Hamza

Nagaraja

(2006). Tsunami: Psychosocial aspects of Andaman and Nicobar islands. Assessments and interventions in the early phase. International Review of Psychiatry, 18, 233–239.

20.

Math

S. B.

John

J. P.

Girimaji

S. C.

Benegal

Sunny

Krishnakanth

Nagaraja

(2008). Comparative study of psychiatric morbidity among the displaced and non-displaced populations in the Andaman and Nicobar Islands following the tsunami. Prehospital and Disaster Medicine, 23, 29–34.

21.

Meyers

(1994). ‘Women and children first’: Introducing a gender strategy into disaster preparedness. Focus on Gender, 2, 14–16.

22.

Morrow

B. H.

(1999). Identifying and mapping community vulnerability. Disaster, 23, 1–18.

23.

Murphy

S. A.

(2010). Women’s and children’s exposure to mass disaster and terrorist attacks. Issues in Mental Health Nursing, 31, 45–53.

24.

Norris

Algeria

(2005). Mental health care for ethnic minority individuals and communities in the aftermath of disasters and mass violence. CNS Spectrums, 10, 132–140.

25.

Norris

Friedman

Watson

(2002). 60,000 disaster victims speak: Part I. An empirical review of the empirical literature, 1981–2001. Psychiatry, 65, 207–239.

26.

North

C. S.

Smith

E. M.

McCool

R. E.

(1989). Acute post-disaster coping and adjustment. Journal of Traumatic Stress, 2, 353–360.

27.

Palinkas

L. A.

Downs

M. A.

Petterson

J. S.

Russell

(1993). Social, cultural, and psychological impacts of the Exxon Valdez oil spill. Human Organization, 52, 1–13.

28.

Perilla

Norris

Lavizzo

(2002). Ethnicity, culture, and disaster response identifying and explaining ethnic differences in PTSD six months after Hurricane Andrew. Journal of Social and Clinical Psychology, 21, 28–45.

29.

Putnam

R. D.

(2000). Bowling alone: The collapse and revival of American community. New York: Simon and Schuster.

30.

Rashid

S. F.

Michaud

(2000). Female adolescents and their sexuality: Notions of honour, shame, purity and pollution during the floods. Disasters, 24, 54–70.

31.

Regehr

Hemsworth

Hill

(2001). Individual predictors of posttraumatic distress: A structural equation model. Canadian Journal of Psychiatry, 46, 156–161.

32.

Richter

Flowers

(2008). Gendered dimensions of disaster care: Critical distinctions in female psychosocial needs, triage, pain assessment, and care. American Journal of Disaster Medicine, 3, 31–37.

33.

Rosborough

Chan

J. L.

Parmar

(2009). Responding to gender-based violence in disasters: Grappling with research methods to clear the way for planning. Disaster Medicine and Public Health Preparedness, 3, 8–10.

34.

Snowden

L. R.

(2003). Bias in mental health assessment and intervention: Theory and evidence. American Journal of Public Health, 93, 239–243.

35.

Somasundaram

D. J.

van de Put

W. A. C. M.

(2006). Management of trauma in special population after a disaster. Journal of Clinical Psychiatry, 67, 64–73.

36.

Summerfield

(1999). A critique of seven assumptions behind psychological trauma programmes in war-affected areas. Social Science and Medicine, 48, 1449–1462.

37.

van Griensven

Chakkraband

M. L. S.

Thienkrua

(2006). Mental health problems among adults in tsunami-affected areas in southern Thailand. Journal of American Medical Association, 296, 537–548.

38.

World Health Organization. (1992a). Psychosocial consequences of disaster: Prevention and management. Geneva, Switzerland: World Health Organization.

39.

World Health Organization. (1992b). ICD-10 classification of mental and behavioural disorders. Geneva, Switzerland: World Health Organization.