Violence against women is strongly associated with suicide attempts: Evidence from the WHO multi-country study on women’s health and domestic violence against women

https://doi.org/10.1016/j.socscimed.2011.05.006Get rights and content

Abstract

Suicidal behaviours are one of the most important contributors to the global burden of disease among women, but little is known about prevalence and modifiable risk factors in low and middle income countries. We use data from the WHO multi-country study on women’s health and domestic violence against women to examine the prevalence of suicidal thoughts and attempts, and relationships between suicide attempts and mental health status, child sexual abuse, partner violence and other variables. Population representative cross-sectional household surveys were conducted from 2000–2003 in 13 provincial (more rural) and city (urban) sites in Brazil, Ethiopia, Japan, Namibia, Peru, Samoa, Serbia, Thailand and Tanzania. 20967 women aged 15–49 years participated. Prevalence of lifetime suicide attempts, lifetime suicidal thoughts, and suicidal thoughts in the past four weeks were calculated, and multivariate logistic regression models were fit to examine factors associated with suicide attempts in each site. Prevalence of lifetime suicide attempts ranged from 0.8% (Tanzania) to 12.0% (Peru city); lifetime thoughts of suicide from 7.2% (Tanzania province) to 29.0% (Peru province), and thoughts in the past four weeks from 1.9% (Serbia) to 13.6% (Peru province). 25–50% of women with suicidal thoughts in the past four weeks had also visited a health worker in that time. The most consistent risk factors for suicide attempts after adjusting for probable common mental health disorders were: intimate partner violence, non-partner physical violence, ever being divorced, separated or widowed, childhood sexual abuse and having a mother who had experienced intimate partner violence. Mental health policies and services must recognise the consistent relationship between violence and suicidality in women in low and middle income countries. Training health sector workers to recognize and respond to the consequences of violence may substantially reduce the health burden associated with suicidal behaviour.

Highlights

► We present data on prevalence and predictors of suicidal behaviour among women from 13 low and middle income settings. ► Prevalence of lifetime suicide attempts ranged from 0.8 to 12.0%. ► Experience of violence predicted suicide attempts across settings, even after accounting for common mental disorders.

Introduction

Durkheim, the original suicide theorist, famously postulated in the 1800s that women were less prone to suicide because they were inclined to reside in the protective domestic sphere and did not trifle in worldly affairs (Kushner & Sterk, 2005). Although subsequent research has challenged Durkheim’s thesis (Canetto and Lester, 1998, Girard, 1993, Kushner and Sterk, 2005, Stark and Flitcraft, 1979, Stark and Flitcraft, 1995), suicidal behaviour among women has received relatively limited attention as a public health concern.

Globally, men are more likely to complete suicide than women, although there is substantial regional variation. In North America and Europe, approximately 3.9 men die from suicide for every one woman, whereas in Asia, sex ratios are more similar at about 1.1 men for every woman (World Health Organization, 2004). In China, women are more likely to complete suicide than men (Phillips, Li, & Zhang, 2002). In contrast, in many settings non-fatal suicidal behaviours are more common among women than men (Beautrais, 2006, Bernal et al., 2007, Bertolote et al., 2005, Nock et al., 2008), but historically have received less attention than fatalities. Indeed, when both mortality and disability are taken into account, the extent of the health burden of suicidality among women becomes clear. Intentional self-inflicted injuries were the 4th leading cause of death and 7th leading cause of DALYS for women aged 15–44 globally in 2005, ranking higher than well-recognised public health issues such road traffic crashes and hearing loss (Ribeiro, Jacobsen, Mathers, & Garcia-Moreno, 2008).

Despite this substantial health burden, little is known about the extent of suicidal behaviour and potentially modifiable risk factors in low and middle income settings (Vijayakumar, John, Pirkis, & Whiteford, 2005). Several recent international studies, including the world mental health surveys (Borges et al., 2007, Gureje et al., 2007, Joe et al., 2008, Nock et al., 2008), WHO SUicide PREvention Multisite Intervention Study on Suicidal behaviours (SUPRE-MISS) (Bertolote et al., 2005, Tran Thi Thanh et al., 2006), along with various in-country secondary school-based adolescent health surveys (Blum et al., 2003, Muula et al., 2007, Rudatsikira et al., 2007a, Rudatsikira et al., 2007b) and other studies (Agoub et al., 2006, Alem et al., 2007b, Kebede and Alem, 2007), have found the reported prevalence of attempted suicide among women and girls to be approximately 1–8%. Unfortunately, most of these studies report only on the role of socio-demographic characteristics and/or mental disorders as risk factors for suicidal behaviour (Bertolote et al., 2005, Nock et al., 2008). Although there is evidence from North America and Europe, and recently from India, confirming that adverse childhood experiences (Dube et al., 2001, Fergusson et al., 2008) and partner violence (Ellsberg, Jansen, Heise, Watts, & Garcia-Moreno, 2008) are associated with suicidal behaviour, only a limited number of studies from low and middle income countries have explored the role of these risk factors (Ahmed et al., 2004, Alem et al., 2007b, Blum et al., 2003, Borges et al., 2008, Maselko and Patel, 2008, Naved and Akhtar, 2008, Pillai et al., 2008, Vizcarra et al., 2004). Early exposure to violence and/or trauma may increase subsequent feelings of depression and affect ability to cope with life stressors, and thus be related to suicidal outcomes (Brodsky & Stanley, 2008, Fergusson et al., 2008). Similarly, there is discussion in the literature about the role of having a dowry/bride price, control over choosing one’s husband, and being childless in marriage in increasing suicide risk, but evidence only from a handful of studies, mainly in Asian settings (Ahmed et al., 2004, Kumar, 2003, Yusuf et al., 2000). Each of these factors may result in poor mental health status directly, but also may be associated with restricted autonomy and loss of control, which can then in turn cause poor mental health status (Canetto & Lester, 1998). Having increased autonomy and social support available may buffer associations between poor mental health status with violence and gender norm variables by mitigating feelings of isolation and hopelessness (for example, De Silva, Huttly, Harpham & Kenward, 2007).

We hypothesized that early childhood and adult experiences of violence, more conservative gender norms and poor adult mental health status would predict increased levels of suicidal outcomes (Fig. 1). This paper uses data from the WHO multi-country study on women’s health and domestic violence against women, which included 20967 women from 9 mainly low and middle income countries. We describe prevalence of suicide outcomes, and assess common and consistent predictors of reported suicide attempts across settings.

Section snippets

Methods

The study methods have been described in detail elsewhere (Garcia-Moreno, Jansen, Heise, & Watts, 2005). Briefly, representative population-based surveys of women aged 15–49 years (aged 18–49 in Japan) were conducted between 2000–2003. In four countries, study sites were more rural (‘province’) and urban (‘city’) locations: Zona da Mata de Pernambuco and Sao Paulo, Brazil; Cusco and Lima, Peru; Nakhonsawan and Bangkok, Thailand; Mbeya and Dar es Salaam, Tanzania. In five countries the survey

Results

A detailed description of the demographic characteristics of the women interviewed is published elsewhere (Ellsberg et al., 2008, Garcia-Moreno et al., 2005). The percentage of women who had ever been in an intimate partnership ranged from 68.5% in Thailand city to 93.9% in Japan city. The most common form of violence reported in most sites was intimate partner violence (70.9% in Ethiopia province to 15.4% in Japan city of ever-partnered women, with 9 of 13 sites above 40%). Non-partner sexual

Discussion

These findings provide new information on the prevalence of suicidality among women and demonstrate a strong association with experiences of violence. There was a 15-fold variation in the prevalence of suicide attempts across sites, 4-fold variation in lifetime suicidal thoughts and a 7-fold variation in recent suicidal thoughts. These findings are consistent with SUPRE-MISS, which documented a 10–14-fold variation in rates of suicidal ideations and attempts in men and women (Bertolote et al.,

Conclusion

National suicide rates have been identified as key indicators of state progress on mental health, along with the existence of national mental health policies and adequate resource allocation (Lancet Global Mental Health Group, 2007). The high prevalence of suicidal thoughts and attempts in this study and the strong association with violence against women illustrates the necessity of prioritizing violence in suicide reduction strategies for women. The recognition of violence as a major risk

Acknowledgements

We gratefully acknowledge the participants in the WHO study, and anonymous reviewers for helpful comments.

References (61)

  • M. Phillips et al.

    Suicide rates in China, 1995–1999

    Lancet

    (2002)
  • P.M. Pronyk et al.

    Effect of a structural intervention for the prevention of intimate-partner violence and HIV in rural South Africa: a cluster randomised trial

    Lancet

    (2006)
  • P.S. Ribeiro et al.

    Priorities for women’s health from the global burden of disease study

    International Journal of Gynecology & Obstetrics

    (2008)
  • S. Saxena et al.

    Global Mental Health 2: resources for mental health: scarcity, inequity and inefficiency

    Lancet

    (2007)
  • H.R. Yusuf et al.

    Injury-related deaths in women aged 10–50 years in Bangladesh, 1996–1997

    Lancet

    (2000)
  • K. Ahmed et al.

    Violent deaths among women of reproductive age in rural Bangladesh

    Social Science & Medicine

    (2004)
  • A. Alem et al.

    Awareness and attitudes of a rural Ethiopian community towards suicidal behaviour: a key informant study in Butajira, Ethiopia

    Acta Psychiatrica Scandanavia

    (2007)
  • A. Alem et al.

    Suicide attempts among adults in Butajira, Ethiopia

    Acta Psychiatrica Scandanavia

    (2007)
  • G. Andrews et al.

    Child sexual abuse

  • A.L. Beautrais

    Women and suicidal behavior

    Crisis

    (2006)
  • J.M. Bertolote et al.

    Suicide attempts, plans, and ideation in culturally diverse sites: the WHO SUPRE-MISS community survey

    Psychological Medicine

    (2005)
  • M. Beusenberg et al.

    A user’s guide to the self-reporting questionnaire (SRQ)

    (1994)
  • R.W. Blum et al.

    Adolescent health in the Caribbean: risk and protective factors

    American Journal of Public Health

    (2003)
  • G. Borges et al.

    Traumatic events and suicide-related outcomes among Mexico City adolescents

    Journal of Child Psychology and Psychiatry

    (2008)
  • G. Borges et al.

    The epidemiology of suicide-related outcomes in Mexico

    Suicide and Life-Threatening Behaviors

    (2007)
  • S.S. Canetto et al.

    Gender, culture and suicidal behavior

    Transcultural Psychiatry

    (1998)
  • S.R. Dube et al.

    Childhood abuse, household dysfunction, and the risk of attempted suicide throughout the life span. Findings from the adverse childhood experiences study

    JAMA

    (2001)
  • C. Garcia-Moreno et al.

    WHO multi-country study on women’s health and domestic violence against women: Intial results on prevalence, health outcomes and women’s responses

    (2005)
  • C. Girard

    Age, gender, and suicide: a cross-national analysis

    American Sociological Review

    (1993)
  • O. Gureje et al.

    The profile and risks of suicidal behaviours in the Nigerian survey of mental health and well-being

    Psychological Medicine

    (2007)
  • Cited by (397)

    View all citing articles on Scopus
    View full text