The Partition of India in 1947 was the start of a long history of dispute between India and Pakistan for control of Kashmir, which today remains divided into three parts governed by India, Pakistan and China. Over the last 20 years, a liberation struggle between India and Kashmiri militants has led to at least 20,000 deaths and 4,000 disappearances in the Indian part of Kashmir [1].

A community survey done by Médecins Sans Frontières in 2005 found high levels of ongoing violence across the region, with civilians caught in the middle. The majority of people surveyed stated having been exposed to crossfire (86%) and round-up raids (83%). High numbers of people reported being subjected to maltreatment (44%), forced labour (33%), kidnapping (17%), torture (13%) and sexual violence (12%). [2]

Exposure to violence has potentially important implications for mental health [3]. This paper presents the findings of the community assessment survey done by Médecins Sans Frontières in 2005. The study, which was done to inform program planning, assessed the mental health and socio-economic impact of the ongoing violence, and the sources of support.

The survey was conducted in mid-2005 in the Indian part of Kashmir (Kupwara and Badgam, totalling 101 villages and a combined population 145,000 people). The methodology is described in detail elsewhere [2]. Briefly, sample size calculation assumed a prevalence of trauma-related psychological problems of 20% [4]; using a precision of 5% (confidence interval 95%) and a design effect of 2, the minimum sample size was estimated at 492. A two-stage cluster sampling design was used to cover 30 villages (randomly selected), resulted in 17 households per village. Within the household participants aged ≥ 18 years were selected randomly. Informed consent was attained for all participants and MSF's independent Ethical Review Board granted ethical approval.

510 of 548 (93%) interviews were completed. Reasons for refusal to participate (25) and stopping the interview (13) included: lack of time, distrust, and being emotionally upset. The average age of respondents was 37.7 years (range 17–90) with an equal gender distribution (males = 53%; 270; p > 0.05). Demographics are described in detail elsewhere. [2]

The data presented in this article were gathered to inform MSF's programme to provide mental health support in Kashmir. Using the SRQ (a tool that has been validated in other Indian studies [6–10]) we found the population had been exposed to high levels of violence [2] which resulted in one third of the respondents suffering from psychological distress and considering suicide. For both genders, currently not feeling safe was associated with psychological distress. For males 'violation of modesty', displacement, and disability were associated with psychological distress while risk factors for females included witnessing killing and torture. Respondents with high psychological distress rated their own health and socio economic functioning as poor. The most common coping mechanism was withdrawal.

Overall, one-third of respondents reported psychological distress. This compares to a prevalence of 36% found in a study done in among Afghan women in a refugee camp [15] using the same instrument and similar cutoff score, but differs substantially from another SRQ study done in a non-conflict area in India [16] where 18% prevalence of psychological distress was found among low-income urban women, using a relatively low cut-off score (7/8). (Using this lower cut-off would have given a prevalence of psychological distress of 71.4%). The contextual difference in these studies – exposure to chronic violence as compared to 'common' stressors of daily life for women in low urban settings – may account for this difference.

The Self Reporting Questionnaire (SRQ) showed that a third of respondents had contemplated suicide. Suicidal thoughts are common for depressive disorders [17] but do not always lead to a suicide attempt. Our findings are in line with a previous study that reported high suicide rates in this region [18]. A high prevalence of suicidal thoughts is more often reported among populations suffering from chronic violence, with a similar prevalence (33%, 96, n = 297) reported in a population of Afghan refugee women in Pakistan using the same questionnaire (SRQ).

In our study women had significantly higher psychological distress than man. This is in line with other studies showing women suffering more from anxiety disorders than men after confrontation with violence [20]. Feeling safe was found in other studies to be an important pre-condition for being able to deal with adverse traumatic experiences [21, 22], and this was also found in our study.

For males, the most important risk factors for developing psychological distress were 'violation of modesty', displacement and disability. It is possible that these experiences are the most distressing because they interfere with the cultural values and roles of males in Kashmir society: upholding their dignity and being able to protect and feed their families. Those who self-experienced 'violation of modesty' had a threefold chance of suffering from psychological distress (p = 0.001). 'Violation of modesty' is regarded as very degrading and in the few studies on male sexual violence is associated with multiple perpetrators and high levels of physical beating [23, 24], which can further contribute to psychological distress.

For women most psychological distress was associated with feelings of powerlessness – dependency on others for daily living, and witnessing killing and torture. Women have lower confrontations with violence, which can be partly explained by their being largely confined to the home [2]. The significant association of witnessing and psychological distress among females may relate to feelings of helplessness and guilt caused by the witnessing may be more traumatic than experiencing the violence themselves.

Both males and females with high levels of psychological distress rated their own health as much poorer compared to those who did not have high levels of psychological distress (male: OR 4.4; female: OR 3.4). Non-specific health complaints have been associated with (traumatic) stress in other studies [25–27]. It is also possible that people do not understand the relationship between physical symptoms and mental stress [28] or have difficulty to articulate their emotional status and use physical symptoms to articulate mental distress [29].

High psychological distress among males was significantly associated with visiting health services more frequently. Increased use of medical services by those suffering from traumatic-stress related problems are common [30, 31], with up to a 25% increase in number of visits to health care facilities reported in other studies [32–34]. We found this relationship in our survey for males, but not for females. This may be explained by the fact that for both cultural and security reasons females depend on male escorts in order to access health services, restricting their movements.

In our population, high psychological distress is associated with substantially increased likelihood of socio-economic dysfunction, and this has been reported in both Western [35, 36] and Asian [15] contexts. Socio-economic dysfunction can have broad implications, for example by reducing capacity of females to give care to the children or for males to generate income (according to traditional roles).

The most common coping mechanisms such as withdrawal (self-isolation, stop speaking) and aggression may also be symptomatic of depression and/or anxiety disorder (including post-traumatic stress disorder, PTSD). Religion and family assistance are mentioned less frequently as sources of support. This is in contrast to a study conducted in Afghanistan that showed religion and reading the Koran as the two main coping mechanisms for two being confronted with violence [15].

The high levels of violence confronted by the Kashmiri population have resulted in high prevalence (33%) of mental health problems. Poor self-rated health and likelihood of poor socio-economic functioning were associated with high levels of psychological distress. Mental health problems in this context of chronic violence should receive full attention through the provision of appropriate community-based services that would improve access to care and reduce the burden on the health system.

The authors declare that they have no competing interests.