34 Prevention of suicidal behaviour in females: Opportunities and obstacles

Females have higher rates of suicidal ideation and behaviour, and lower rates of suicide mortality than males. This is a dominant, but not a universal pattern, both across and within countries. One of few national exceptions is China, a country where both non-fatal and fatal suicidal behaviours are most common in women. There is also cultural heterogeneity in meanings of, and attitudes about female suicidal behaviour. In some cultures, suicide is viewed more negatively in women, while in other cultures in men. This cultural diversity in gender patterns and meanings of suicidal behaviour challenges essentialist perspectives on female suicidal behaviour, and calls for culture and gender-grounded theory, research and prevention. The perception, dominant in industrialized countries, that suicide is a male behaviour is a challenge, but can also be an opportunity in prevention in that it may discourage female suicide. Evidence indicates that critical to the prevention of female suicidal behaviour is attention to social, economic and political factors, including structures of social inequality.

Suicidal behaviour is a significant problem for girls and women around the world. In countries where the epidemiology of suicidal behaviours is recorded, females have higher rates of suicidal ideation and behaviour than males. Suicide mortality tends to be lower in females than in males, though this pattern is not universal (Canetto 2005). The problem of female suicidality, however, is under-regarded and under-studied (Beautrais 2006).

In this chapter I focus on female suicidal behaviour around the world. First, I address the cultural variability in definitions and recording practices of suicidal behaviour. This is to note the limitations of comparing epidemiological patterns across cultures, and the difficulty of theorizing based on cross-cultural data. Next, I review epidemiological trends in female suicidal behaviour. I then present and discuss theories and research on female suicidal behaviour from around the world. Following that, I provide a summary of the international evidence on female suicidal behaviour, as well as highlights of a theory building on these global data. In the last section, I address the implications of our current global knowledge for the prevention of female suicidal behaviour, with attention to obstacles and opportunities.

A fundamental issue when examining suicidal behaviour across cultures is how suicidal behaviour is defined. This is because what one culture considers critical for an act to be labelled suicidal may not coincide with what another culture considers essential.

In industrialized countries suicide is defined as deliberately self-inflicted death. In other countries, however, it is not necessary for a death to be self-induced to be considered a suicide. One example is the case of ritual killing of older widows by male kin among the Lusi in the Kaliai district of Papua New Guinea. The Lusi view the ritual killing of widows as suicides, since widows presumably demand to be killed to avoid becoming dependent on their children. By contrast, the German and Australian local authorities consider widow-killing a murder (Counts 1980, 1984). It is noteworthy that widow-killing by kin does not reflect a Lusi attitude about widowhood in general. It is an attitude about female widowhood. Among the Lusi, widowers are not subject to ritual killing by kin.

In industrialized countries suicide is generally assumed to involve individual choice. Suicide voluntariness may, however, be difficult to establish. Consider, for example, the case of sati, a mode of death that has been outlawed, but is not extinct in modern India. Sati involves a widow climbing on the funeral pyre of the deceased husband to be burned with him. Sati is presumably voluntary, which would make it a suicide, at least by industrialized countries' standards. The meanings and social consequences of widowhood and sati do, however, raise questions about choice in sati. According to Hindu tradition, if a husband dies before his wife, it is because of a wrong the wife committed. In traditional communities a Hindu widow is expected to submit to social and economic restrictions, ranging from exclusion from festive events to the prohibition to marry, to banishment without possessions to an ashram, a widows' house. By contrast, a Hindu widow who dies by sati is thought to bring great fortune to herself and her kinship. Thus, in Hindu tradition, a widow does not have ‘good life’ choices, only a ‘good suicide’ choice (Andriolo 1998; Cheng and Lee 2000).

The complexity in the determination of a behaviour as suicidal is further illustrated by cases of women's deaths by domestic burning. These deaths, which typically result from having caught fire while around a household open stove, have been reported in India, Sri Lanka, Iran, South Africa and Zimbabwe. What makes these suicides questionable is that they typically occur in the context of dowry or other disputes with in-laws. Even those deaths that appear and are recorded as suicides or accidents may be murders, either directly perpetrated or indirectly triggered through psychological pressure (Waters 1999; Mzezewa et al. 2000; Laloe and Ganesan 2002; Sukhai et al. 2002; Batra 2003; Groohi et al. 2003; Kumar 2003; Aaron et al. 2004; Laloe 2004; Maghsoudi et al. 2004; Mohanty et al. 2005; Lari et al. 2007; Rastegar et al. 2007).

In conclusion, the determination of a behaviour as suicidal is influenced by cultural and political factors. Under the influence of cultural and political factors, female homicides may be recorded as suicides, while female suicides may be registered as accidents or undetermined deaths. These cultural and political influences are a liability in that they introduce unaccounted variance in epidemiological data. They are also an oppotunity in that they challenge, and ultimately expand frameworks for understanding suicidal behaviour.

In countries where the epidemiology of suicidal behaviours is recorded, females tend to have higher rates of suicidal ideation and behaviour than males. This is a common, but not a universal pattern. One exception is Finland, where women and men have similar rates of nonfatal suicidal behaviour (Schmidtke et al. 1996). A caveat with regard to the data on suicidal ideation and non-fatal suicidal behaviour is that they come from selected sources and communities that is, mostly from hospital records, urban-areas and industrialized countries.

According to World Health Organization (WHO) records, suicide mortality is lower in females than in males, though this pattern is also not universal (2007). Specifically, suicide is most frequent in women in China, a country with 21 per cent of the world's population and 56 per cent of the world's female suicide (Murray and Lopez 1996). A strength of the suicide data is that they are collected on a national basis and reported to the WHO by approximately 130 of the world's 192 nations recognized by the United Nations (Bertolote 2001). A limitation is that the national suicide mortality reported to the WHO come from a selected group of countries, with industrialized countries being over-represented in the sample (Vijayakumar et al. 2005b).

The over-representation of females among the suicidal, and their under-representation among the dead by suicide is a paradox. One would expect the group with the highest number of life-threatening acts to also have the highest mortality from those acts (Canetto and Sakinofsky 1998). The gender paradox of suicidal behaviour is particularly common in industrialized countries, where it is often taken as a manifestation of essential and stable female–male differences. However, the gender paradox of suicidal behaviour is not consistently found even in industrialized countries. For example, in the US, there are exceptions to the gender paradox of suicidal behaviour when one examines female–male suicidality patterns by age, ethnicity or region. US exceptions are the similar rates of non-fatal suicidal behaviour recorded among female and male adolescents of Native Hawaiian descent (Yuen et al. 1996) as well as among female and male Pueblo Indian adolescents (Howard-Pitney et al. 1991). Exceptions to the gender paradox of suicidal behaviour are also found in Central American, South American and Asian countries. For example, in Brazil, Cuba, the Dominican Republic, Ecuador, Hong Kong, Paraguay, the Philippines, Singapore, and Thailand, young females' suicide mortality exceeds that of young males (Canetto and Lester 1995a).

It is also noteworthy that the gender paradox of suicidal behaviour is not always a stable pattern. Historical analyses of suicide mortality trends show that in some of the countries where the gender paradox had been documented, gender patterns of suicide mortality have been shifting. For example, in Denmark the gender gap in suicide mortality has been narrowing (Canetto and Sakinofsky 1998).

In conclusion, the diversity in gender patterns of suicidal behaviours within and between countries as well as across time, combined with the selectivity of the data on suicidal behaviours, suggest that the gender paradox of suicidal behaviour does not represent a universal and fixed female–male difference in suicidal behaviour, as industrialized countries' suicidologists have often assumed (Canetto 2005). Rather, the gender paradox of suicidal behaviour is a culturally and historically specific phenomenon. When one takes a global and historical perspective, one finds that there is not a distinctly female (or a uniquely male) way ‘to do suicide’. Rather, female and male patterns of suicidal behaviour are in some cases similar, and in other cases different, depending on cultural context and time period (Kushner 1995; Canetto and Lester 1998).

Internationally influential theories and research on suicidal behaviour are generated mostly in industrialized countries. As a result, these theories and research tend to focus on industrialized countries' own dominant epidemiological gender pattern—that is, females' high suicidal morbidity and their low suicidal mortality. As noted earlier, this gender pattern is common, but neither universal nor stable, even within industrialized countries. This means that industrialized countries' dominant theories and findings are culture- and time-specific, though they often purport to be universally relevant (Canetto 2005). Given their global influence, these theories and research, however, deserve attention. This section opens with a review of dominant, industrialized countries-focused theories of female suicidal behaviour. These theories are critiqued in light of evidence from both industrialized, emerging and developing countries.

A most notable feature of industrialized countries' theories and research on suicidal behaviour is their under-regard for female suicidality. Simply put, dominant theories and research mostly ignore female suicidal behaviour (Beautrais 2006).

A factor contributing to the under-investment in female suicidal behaviour is females' lower rates of suicide mortality, relative to males. The paradox is that females have much higher rates of suicidal ideation and non-fatal suicidal behaviour than males. In fact, when morbidity and mortality from suicidal behaviour are considered together, females emerge as the most suicidal group (Beautrais 2006).

Another factor likely contributing to industrialized countries' limited attention to female suicidal behaviour is a long-standing tradition of conceptualizing suicide as a male behaviour (see Kushner 1993, 1995 for reviews). Many industrialized countries' authoritative theorists, including Durkheim (1897/1951), argued that suicide requires a degree of courage and intelligence they believed could be found only in men. Specifically, according to Durkheim, women are too timid, too weak, too conformist, and too dull to kill themselves. According to this theory, women are immune to suicide—at least as long as they act ‘like women’, that is as long as they stay subordinate to men and subsumed within ‘traditional’ institutions, such as ‘traditional’ marriage. Women who suicide, Dublin claimed, must have experienced a ‘marked increase in … schooling and employment. … Greater economic and social independence … played a role’ (Dublin 1963). This perspective also assumes that when women act ‘masculine’, that is when they venture into such masculine domains and activities as education and employment, they risk becoming suicide casualties, like men.

A major theme in dominant industrialized countries' theories and research is that female suicidal behaviour is an expression of individual pathology. Females are thought to be naturally protected from suicide—since suicide is considered masculine behaviour. Therefore, suicidal women who kill themselves, are believed to be especially abnormal and unfeminine (Canetto 1997a; Deluty 1988–1989; Kushner 1995).

In industrialized countries there has been continuous interest in the role of female-specific biological factors in the risk for, and protection from suicidal behaviour—an interest that contrasts with the consistent lack of concern for male-specific biological factors (Canetto 1995b, 1997a). The evidence for female-specific biological factors in suicidal behaviour is, however, at best equivocal. For example, some studies from industrialized countries find pregnancy and motherhood to be protective against suicidality (Lindahl et al. 2005; Stallones et al. 2007). At the same time, pregnancy and motherhood are less protective when pregnancy is unwanted or the mother is younger than 20 (Appleby 1991; Gissler and Lonnqvist 1996; Vaiva et al. 1997). There is also evidence, from some developing countries (e.g., China, India, Kuwait), of an association between female suicidal behaviour and childlessness (Wolf 1975; Batra 2003; Fido and Zahid 2004). However, in these developing countries, mothers are also at risk for suicidal behaviour if they have girls, the culturally devalued children (Wolf 1975; Pearson 1995; Waters 1999; Ji et al. 2001; Meng 2002; Batra 2003). Together, these findings suggest that what may be protective about motherhood are not some universal biological processes, but rather, its social meanings, when positive.

In industrialized countries, much attention has been given to the role of mental disorders in female suicidality, particularly depression (Canetto 1997a). This emphasis on female psychopathology fits the dominant industrialized countries discourse on suicide. In industrialized countries, mental disorders are considered critical precursors of suicidal behaviour. By contrast, in developing countries, mental disorders are not viewed as significant in suicidal behaviour, with external stressors being believed to be most influential (Marecek 1998; Vijayakumar et al. 2005b,). Industrialized countries' psychopathology perspective is consistent with their mental disorders epidemiology trends. In these countries, girls and women are more likely to be diagnosed with the mental disorders (e.g., depression) considered precursors of suicidal behaviour (Canetto 1997b). Industrialized countries' psychopathology perspective on female suicidal behaviour has many limitations, a key one being that it does not explain their own gender paradox of suicidal behaviour. If suicidal behaviour were simply a function of psychopathology, females, the group with the highest rates of precursor mental disorders and of non-fatal suicidal behaviour, should also have highest rates of suicide mortality, which is not typically the case.

Female suicidal behaviour is often explained as a response to interpersonal problems (Canetto1997a). This theory is prevalent in industrialized countries and in developing countries, with the difference being that in developing countries, interpersonal problems are often viewed as critical to male suicidal behaviour as well (Canetto and Lester 1998; Marecek 1998; Canetto 2005). In industrialized countries, suicidal women are believed to be psychologically impaired, and to break into suicidal behaviour in response to trivial, private interpersonal problems. By contrast, in these countries, male suicide is constructed as a decision in response to important impersonal adversities and losses, such as work problems or illnesses (Canetto 1992–1993; Canetto 1997a). Women's suicidal behaviour is also often assumed to be impulsive, ‘not serious’ and ‘manipulative’, as compared to men's suicidal behaviour, which is presumed to be planned, ‘serious’, and death-aiming (Canetto 1997a).

Research does not support these assumptions. First, while there is evidence that interpersonal factors play a role in female suicidal behaviour, it does not appear that suicidal behaviour is more interpersonally driven in women than in men. For women, as well as for men, both interpersonal and impersonal factors appear significant in suicidal behaviour (Canetto and Lester 1995b, 2002; Hjelmeland et al. 2002b). For example, interpersonal difficulties emerge as the dominant issues in women and men's suicide notes (Canetto and Lester 2002). Also, evidence shows that employment protects both women and men from suicidal behaviour (Kposowa 2001; Stallones et al. 2007).

Second, the kind of interpersonal problems associated with female suicidal behaviour are anything but trivial. Studies from both industrialized as well as developing countries find that rates of suicidal behaviour are higher in women who have experienced abuse. This abuse, which may be sexual, physical and/or psychological, often takes place within family relationships (Counts 1980, 1984, 1987; Stark and Flitcraft 1995; Ji et al. 2001; Batra 2003; Fergusson et al. 2005; Verona et al. 2005; Roy and Janal 2006,).

Third, research findings challenge the idea that the interpersonal problems associated with female suicidal behaviour represent ‘simply’ private, domestic matters. Studies from developing countries are particularly enlightening in this regard. Some argue that for single young women, suicidality is associated with private turmoil, instability, and family conflict. However, for young single suicidal women from non-industrialized societies (e.g., in East Kwaio, Malaita, in the Solomon Islands, or among the Aguaruna of the Peruvian Amazon), family conflict typically revolves around culturally-based restrictions on women's self-determination and mobility, including a culturally-specific sexual impropriety, the choice of marriage partner, or interest in education (Akin 1985; Brown 1986). Under these circumstances, the young woman's suicide is viewed as culturally understandable, if not expected. Consider next how, in several developing countries, suicidality among married women is frequently associated with conflict with, and abuse from intimate partners and/or in-laws. For example, in rural China, conflict with, and abuse from in-laws often follows childlessness or the failure to produce a son, a situation revealing a cultural tradition of female devaluation and oppression (Wolf 1975; Ji et al. 2001). These married women's vulnerability to abuse by their husbands and in-laws is enabled by cultural traditions, including patrilocal marriages and patrilineal inheritance traditions. Having moved away from their natal village uneducated and with no personal assets, married women end up isolated and vulnerable. Female suicide in response to family conflict and abuse is culturally sanctioned behaviour in these societies. It is a way for women to obtain justice against their abusers, though only post-mortem. For example, in China, a married woman's suicide may oblige her in-laws to provide her kin with financial compensation (Wolf 1975; Pearson 1995; Pritchard 1996; Ji et al. 2001; Meng 2002; Pearson and Liu 2002). Finally, suicidality in older women is also often based in cultural practices of female devaluation and oppression, though not necessarily in family conflict. In communities where older women have high suicide rates, one frequently finds the belief that older women, particularly older widows, are a burden, with suicide being considered acceptable or even expected of widows. An example is Papua New Guinea's ritual killing of widows, a practice which the Lusi considered suicide (Counts 1980, 1984). What these examples illustrate is that the presumed private relationship problems associated with female suicidality, including abuse by family members, are often grounded in, and supported by cultural traditions. Another important point highlighted by these examples is how suicidal behaviour is culturally grounded. In each culture, there are circumstances when suicidal behaviour is socially understandable, and even expected (Healey 1979; Counts 1980; Heshusius 1980; Johnson 1981; Counts 1984; Canetto 1997a; Counts 1987; Ji et al. 2001; Canetto 2005). As noted by Kroeber, ‘culture[s] not only define[s] certain situations that call for suicide but often indicate[s] the correct way to execute it’ (1948, cited in Jeffreys 1952, p. 118). The fact that one can better see the influence of culture in foreign suicidal scenarios than in one's own is an indication of the difficulty of stepping out of one's own cultural habits and norms.

Finally, research does not support the assumption that female suicidal behaviour is more impulsive and less ‘serious’ than male suicidal behaviour. Industrialized countries studies of impulsivity, motives and intent indicate that suicidal females and males may be more similar than different in these domains, with the wish to die being the dominant intent reported by all (Hjelmeland et al. 2000; Hjelmeland et al. 2002a). What then may account for women's lower rates of suicide mortality? We know that the fatality of a suicidal act is a function of several factors beside intent, including the suicide method, the place of the suicide, the likelihood to be quickly found, and the likelihood of effective care. The more rapidly-lethal the method, and the less public the suicidal act, the lower the likelihood of survival. It has been speculated that women's generally lower suicide mortality is a function of their being less likely to use immediately-lethal suicide methods. In industrialized countries, suicide by firearms, a method more commonly used by men, carries higher fatality potential than poisoning, a method more commonly used by women, hence men's higher suicide mortality than women's, despite their possibly similar intent. By contrast, in developing countries, suicide by poisoning, even one that is carried out in a public area, carries a high fatality potential due to the toxicity of the poisons used (e.g. agrochemicals), and the likelihood that effective medical care may not be available quickly enough to make a difference. At the same time, it is important to remember that the choice of suicide method as well as the outcome of the suicidal act are influenced by cultural norms, that is, by what cultures consider permissible method and outcome for particular persons, given their personal characteristic and situation. The fact that firearms are the dominant suicide method for males, and that most males die of suicide in the US, are likely influenced by the association of firearms, suicide and masculinity in US culture. Similarly, in the US, non-fatal female suicidality takes place in a cultural context in which ‘attempted’ suicide is considered more feminine and appropriate for women (Canetto 1997a; Canetto and Lester 1998; Canetto 2005).

In this chapter, I examined women's suicidal behaviour from a global perspective. This examination was constrained by the cultural diversity in definitions of suicidal behaviour. Naming and registering an act as suicidal is a cultural and political act.

Studies reveal suicidal behaviour to be a significant problem for girls and women around the world. When morbidity and mortality are considered together, girls and women bear the greatest burden of suicidality. Research also shows that a common risk factor for female suicidal behaviour is the experience of abuse.

Global evidence challenges widespread assumptions about women and suicidal behaviour. One such assumption, based on Durkheim's (1897/1951) theory, is that women are immune from suicide as long as they remain ‘feminine’, family-bound, and socially subordinate. In fact, studies show that membership in tightly structured social units, especially patriarchal families, is a risk factor for female suicidal behaviour in some cultures (Ji et al. 2001; Altindag et al. 2005; Kushner and Sterk 2005; Vijayakumar et al. 2005a; van Bergen et al. 2006). There is also evidence that, contrary to dominant theory, individualism and social equality protect women from suicide, especially among the young. A recent study of thirty-three developing and industrialized countries found that women's suicide rates were lower in countries with social structures emphasizing individualism and social equality (Rudmin et al. 2003).

Global evidence also highlights the heterogeneity in women's risk for, and typical forms of suicidal behaviour across and within cultures. For example, female suicide mortality is higher than male suicide mortality in China, but lower than male suicide mortality in the US. Also, among European Americans, female adolescents have high rates of non-fatal suicidal behaviour than males. However, among Native Hawaiians, female and male adolescents have similar rates of nonfatal suicidal behaviour. Furthermore, there is heterogeneity within and across cultures in the conditions, meanings and consequences of female suicidal behaviour. For example, in some cultures suicide is viewed more negatively in women, while in other cultures in men.

This review has shown that dominant, industrialized-countries theories have disregarded female suicidality and also ignored the cultural variability in female suicidal behaviour. To explain women's suicidal behaviour across cultures, theory needs to start from a foundation on gender and culture. An example of such approach is cultural scripts theory (Canetto 1997a, 1997b, Canetto and Lester 1998, Canetto 2005). Cultural scripts can be thought of as norms of suicidal behaviour, that is, beliefs about, and practices of suicidal behaviour. A cultural script defines the meanings of suicidal behaviour, and the conditions under which suicidal behaviour is relatively acceptable, or even expected. A cultural script has also to do with the suicidal scenario, that is the events leading to the suicidal behaviour, the protagonists of the suicidal situation, the emotions and motives associated with the suicidal behaviour, and the suicide method. Furthermore, cultural scripts guide community responses to the suicidal behaviour, including those of experts and authorities.

Cultural script theory is grounded on the observation of a correspondence between scripts of suicidal behaviour and patterns of suicidal behaviour. Ideology and epidemiology likely reinforce each other. The fact that in some countries females are less likely to die of suicide likely contributes to the belief in the masculinity of suicide. This belief, in turn, can act as a social norm and a model, discouraging female suicide. The belief in the masculinity of suicide may also contribute to the under-recording of female deaths as suicide, especially when they are ambiguous, further perpetuating the norm and belief in the masculinity of suicide (Kushner 1993, 1995; Canetto and Sakinofsky 1998).

According to this theory, the risk for suicidal behaviour needs to be evaluated in light of local cultural scripts. This is based on the observation that those most at risk for suicidal behaviour are not necessarily the individuals who are most psychologically impaired, or the persons facing the greatest adversities. Rather the persons who engage in suicidal behaviour are those whose characteristics and circumstances also fit local scripts of suicidal behaviour. In other words, suicidal behaviour is most likely when it is socially expected. This does not mean that there is minimal individual agency in suicidal behaviour. It only means that individual agency is articulated in relation to cultural scripts (Canetto 2005). Individuals draw on local scripts in defining their suicidal action and in giving their suicidal behaviour public significance (Rubinstein 1992; Canetto 2005)

Cultural script theory overcomes the limitations of a purely individual perspective on female suicidal behaviour, including industrialized countries conceptualization of suicide as individual psychopathology. Cultural script theory also overcomes the limitations of pure ecological perspectives on suicidal behaviour, such as those dominant in developing countries. By focusing on meanings of gender and suicidality, cultural script theory integrates individual and ecological influences, and accounts for the apparent paradox that those who are most oppressed are not always the most suicidal. Consider, for example, that in the US, suicide rates are highest among older men of European-American descent, and lowest among older women of African-American descent. According to cultural script theory, this is because suicidal behaviour is most likely when it is the culturally-supported response (Canetto 2005).

What are the suicide prevention directions suggested by the global data on female suicidal behaviour? These data challenge essentialist perspectives on female suicidal behaviour. They also call attention to both environmental and individual factors in female suicidal behaviour. Finally, they highlight the importance of attending to culture and gender in prevention work. A culture- and gender-grounded suicide prevention approach means avoiding one size fits all programmes. This approach may take the form of a programme exploring, and educating about local scripts of suicidal behaviour. In a country like the US, where suicide ‘attempts’ are associated with femininity, educational programmes might challenge the notion that non-fatal suicidal behaviour is a feminine way to cope with problems. This notion is not only problematic for females in that it may inadvertently encourage female non-fatal suicidal behaviour. It is also dysfunctional for males in that it may push suicidal males toward killing themselves. As argued by Linehan, ‘due to social pressures against attempted suicide, males … [might] “skip” over the less drastic solution of attempting suicide and go directly to suicide’ (Linehan 1973, pp. 31–32).

There are many obstacles to the prevention of female suicidal behaviour. A major one is the belief, dominant in industrialized countries, that suicide is a male problem. This belief contributes to under-evaluating the severity of the problem of female suicidality. It also obscures the variability in female suicide mortality between and across countries. Educating about females' significant involvement in suicidal behaviour may draw needed attention to it. Studies of female suicidal behaviour will not only provide critical information about female suicidal behaviour, but likely also promote greater understanding of male suicidal behaviour, as issues relevant to both that most readily emerge through the study of women may have been overlooked in studies of men (Canetto 1995a). Education on the seriousness of female suicidal behaviour problem may also improve the care of suicidal females. Studies indicate that care-providers are particularly unhelpful and hostile toward suicidal females out of the misguided view that their suicidality is not serious (see Canetto 1995b for a review).

The belief in females' immunity to suicide is an obstacle, but can also be a resource in the prevention of female suicidal behaviour. Women who believe in the masculinity of suicide may be more reluctant to kill themselves. When working with suicidal women, caregivers can reframe the false negative belief of the masculinity of suicide into an empowering positive belief. Reinforcing women's rejection of suicide as a solution to their problems as well as women's expectation of efficacy in coping with suicidal ideation may support actual effective coping.

Yet, another obstacle to effective suicide prevention in females is the over-emphasis on individual factors, and the under-appreciation of social, economic and cultural factors in female suicidal behaviour. Studies suggest that significant progress in the prevention of female suicidal behaviour will require a paradigmatic shift of focus, from individual to ecological factors (Douglas 1967; Counts 1987; Canetto and Lester 1995a; Corin 1996; Canetto 1997a, b; Canetto and Lester 1998; Waters 1999; Ji et al. 2001; Meng 2002; Pearson and Liu 2002; Laloe 2004; Canetto 2005; Rudmin et al. 2003; Mitra and Singh 2007). Critical in this regard is attention to the role of social equality in females' protection against suicidal behaviour.