This is the first in a series of papers in APT to be devoted to issues of gender and mental health (Bartlett & Hassell, 2001; Cremona & Etchegoyen, 2001; Kennedy, 2001, this issue; Kohen, 2001b; Kohen & Arnold, 2001; see also Kohen, 2001a, this issue). Ideas for further papers are welcome: please write to Dr Dora Kohen or Gillian Blease at the Royal College of Psychiatrists, 17 Belgrave Square, London SW1X 8PG.
Women patients suffer from a range of mental disorders similar to those that men may experience. However, there are some striking differences in the prevalence of specific disorders, and in their presentation and management. Some mental illnesses only occur in women. It seems that women patients may have a different experience of treatment, a consequence of differences in their needs and also of the way that health professionals perceive those needs. These differences are embedded in the wider cultural milieu in which we live. There are particular issues for women patients in relation to, for example, childhood sexual abuse, rape and domestic violence. At present, tools to measure needs of individual patients are generally not gender specific.
Gender differences in the prevalence of mental illness
There are significant gender differences in the prevalence of specific psychiatric disorders, with some disorders more common in women, and others more common in men. The Office of Population Censuses and Surveys (OPCS) survey of 10 000 adults living in private households in the UK found that women were more likely than men to suffer from a neurotic health problem, and men were three times more likely than women to suffer from alcohol dependence and twice as likely to suffer from drug dependence (Meltzer et al, 1995). This gender difference is most marked for eating disorders: over 90% of patients with anorexia nervosa are female.
Similar findings came from the Epidemiological Catchment Area Programme, a community survey of 10 000 individuals over three sites in the USA (Robins et al, 1984). If we include alcohol misuse, drug misuse and antisocial personality disorder – all of which are more common in men – the overall prevalence of mental illness in men and women is very similar.
Other surveys have focused on adults with severe mental illness in particular areas. For example, the PRiSM Psychosis Study in south London, which aimed to identify all people with a psychotic illness living in two particular areas, reported an equal split between men and women (Thornicroft et al, 1998).
Perinatal mental illness
Women also experience specific times in relation to their life cycle when they are at increased risk of developing particular disorders. These are included in ICD–10 (World Health Organization, 1992) under Chapter XV, ‘Pregnancy, Childbirth and the Puerperium’ (O00–O99); O99.3 refers to mental disorders and diseases of the nervous system complicating pregnancy, childbirth and the puerperium and includes the conditions in F00–F99. In clinical practice, we refer to the conditions maternity blues, postnatal depression and post-partum psychosis.
Depression during the first 6 weeks after childbirth in women not suffering from depression before the delivery was first described 30 years ago. Subsequent research has challenged the concept that rates of depression after childbirth are higher than in women with young children outside the postnatal period. It is likely that there are two different populations of women who become depressed after childbirth: one group who only experience depression in relation to motherhood and its demands, and another who are no more likely to become depressed after childbirth than at other times in their lives (Pound & Abel, 1996). Interestingly, the suicide rate in the year following childbirth is one-sixth the rate for a matched control group (Appleby, 1991). In primiparous women, there may be up to a 35-fold increased risk of developing a psychotic illness and needing hospital admission within the first month of childbirth (Kendell et al, 1987), one of the most striking epidemiological findings in psychiatry.
Gender differences in presentation and course of mental illness
Within specific disorders, there may be significant differences in the presentation and outcome for women and men. For example, in schizophrenia, men appear to have a more severe form of the illness, characterised by an early age of onset, poor premorbid adjustment, both typical and negative symptoms and poor outcome, while in women the onset is later and a more affective component is apparent (Castle & Murray, 1991). Among patients with a dual diagnosis admitted for treatment, women seem to have more social contact problems and fewer legal difficulties but more problems with victimisation and medical illness (Brunette & Drake, 1997).
Explanatory models for gender differences
Given these differences in the prevalence of mental illness between men and women and, for specific disorders, the differences in presentation and course, it is helpful for us in trying to understand the needs of women patients if we know some of the reasons for them. Social factors are particularly important in the causation of neurotic disorders, and the repeated finding of higher rates of depression and anxiety in women has encouraged research into possible social causes as an explanation for gender differences in mental illness (Johnson & Buszewicz, 1996).
A possible confounding factor could be differences in the way men and women express mental distress and interact with services, and how health professionals assess and manage male and female patients.
Biological differences between men and women might also play a role; for example, Castle & Murray (1991) attribute gender differences in the presentation and course of schizophrenia to differences in aetiology – men are more likely to have a neurodevelopmental form of the illness, while in women there is a stronger genetic component. More generally, there are likely to be variations in response to drug treatment owing to hormonal differences, for example, oestrogen can potentiate the neuroleptic effects of antipsychotics, so women may require a lower dose than men both in the acute and maintenance phases of therapy (Gold, 1998).
Gender differences in interactions with services
According to Goldberg & Huxley's (1992) ‘filters of care’ framework, women are represented in different proportions at different levels of services. At level two (total morbidity in primary care), women present more to their general practitioner (GP) both with physical and psychological complaints, but GPs refer a higher proportion of men to secondary care services (Johnson & Buszewicz, 1996). Women with high levels of depressive symptoms may be reluctant to seek help from mental health professionals, particularly in deprived areas (Jimenez et al, 1997).
Within mental health services, women may be treated differently: for example, gender differences are reported in the use of the Mental Health Act (MHA) 1983. Over the 10 years from 1988/89 to 1998/99, the number of patients in hospital under a section of the MHA increased, but at a slower rate in women. Although previously there were more detained female patients, since 1991/92, for the first time, the numbers have been higher for men, with the proportion of male detainees rising to 53% in 1998/99 (Department of Health, 2000). In continuing care services for patients with a severe mental illness, older women may receive less intensive input (Perkins & Rowland, 1991). Among those in high contact with the service, men and women were functioning at a similar level, but Perkins & Rowland reported a higher proportion of women in day care facilities designed for people functioning at a lower level.
There may also be significant gaps in services available for women patients. For example, more specialised services for patients with an eating disorder, most of whom are women, are needed. Women who have experienced physical or sexual abuse have particular difficulties if admitted to mixed-sex wards, and few general psychiatric wards have developed adequate women-only areas. Although few women patients require a forensic bed, it is very difficult to find a bed in a medium secure unit for a woman patient.
There may also be gender differences in the diagnosis of mental illness by health professionals. For example, in a sample that met standardised criteria for depression, doctors were more likely to diagnose depression in women than in men (Potts et al, 1991). Looking at personality disorder diagnoses, a case vignette study showed that clinicians used the diagnosis narcissistic personality disorder more in men and histrionic personality disorder more in women (Adler et al, 1990). The gender of the clinician did not appear to affect the process in either case.
Gender differences within the social context
Gender, deprivation and social support
The relationship between poor physical health, impaired psychological functioning and deprived socio-economic background is well documented (Gomm, 1996). Women who display self-neglect and self-harming behaviours are exposed to additional problems when trying to secure accommodation, training and employment. Policy on health in the UK has recently given priority to the need to address the close relationship between inequalities in health and socio-economic circumstances. Combating the effects of these inequalities has become the responsibility of health authorities, with local action channelled through health improvement programmes. The 1992 Health and Lifestyles Survey found that community involvement and social support vary with socio-economic status. Women's health appeared more strongly associated with the social environment than men's (Cooper et al, 1999).
The quality and number of roles individuals play are important, with evidence that those with the fewest family, friendship, working and community roles have the poorest psychosocial health. More frequent and higher-quality social relationships in women can combat the effects of stress. However, individuals vary and we should avoid over-simplified factors for good mental health such as marriage and occupation (Pollock & West, 1987). For women, being part of an emotionally and economically rewarding social network would appear to be helpful in determining mental well-being, unless the demands upon the individual create emotional overload.
Women and families
Family structures across Western Europe are changing, with cohabitation, divorce and re-marriage increasing. Preliminary findings from the 1998 General Household Survey show that 38% of births were outside marriage compared with 26% 10 years earlier, although 61% of the 1998 births were registered jointly by parents living at the same address. Between 1971 and 1990, the number of divorced and separated women with children rose from 290 000 to 650 000, and the number of lone mothers from 90 000 to 390 000. In the late 1980s, European Union figures showed the UK to have a lone parent family rate of 17%, which is among the highest in Europe (Millar, 1992).
Women as parents
Parenting plays an important role, first, in protecting a child from harm and promoting physical and emotional health; second, in setting and enforcing boundaries to ensure the child's and others' safety; and, third, in optimising the child's potential. Parenting can act as a buffer against adversity such as poverty or delinquent influences, but may also be a mediator of damage, as in child abuse. A reasonable consensus exists about ‘bad parenting’, but there is no agreement about its opposite. Although the job is complex and demanding, help is fragmented between different services and reactive in nature, doing little to involve the parent or take preventive action (Hoghughi, 1998).
Mothers who themselves have had poor experiences of parenting may be the least likely to receive practical help with child care and social support to help them function in their new role (Pound & Abel, 1996). Lone teenage mothers are an especially vulnerable group. Women threatened with having their children taken into care may distrust services and feel inhibited from seeking help. A significant proportion of women with a severe mental illness have children, but health professionals tend to ignore these women's role as parents, considering parenting as a social services problem rather than a health issue (Nicholson et al, 1993).
Women's physical health
Women with a mental illness may have less awareness of contraceptive needs, resulting in an increased risk of unwanted pregnancy. They may also be at increased risk of developing a sexually transmitted disease, more specifically having less knowledge of AIDS (Aruffo et al, 1990; Coverdale et al, 1997).
Women and abuse
There has been increasing awareness of the impact of violence against women, in particular, the effects of childhood sexual abuse, domestic violence and rape.
Childhood sexual abuse
Women who have experienced childhood sexual abuse are more likely to suffer social, interpersonal and sexual difficulties in adult life. They seem to have particular problems with intimate relationships, owing to difficulties with trust and a perception of their partners as uncaring and overcontrolling. Abuse may also correlate with an increased risk for a range of mental health problems (Mullen et al, 1994).
A high proportion of women attending accident and emergency departments report a history of domestic violence, and in this group there is a high level of mental health problems. A history of childhood abuse increases a woman's risk of subsequent mental health problems if she is also abused as an adult. In this double-abuse group, there may also be an increased risk of substance misuse (Roberts et al, 1998).
Victims of completed rape are at increased risk of suicide attempts and of having a depressive illness. In addition, perceptions of life threat and actual injury increase the risk of post-traumatic stress disorder (PTSD) (Mezey & Stanko, 1996).
Housing needs and gender
Poor housing conditions are associated with poor health and emotional distress. Risk factors include severe overcrowding, a lack of amenities, noise, environmental hazards and insecure tenure. The council housing sector has been reduced to statutory provision to cover priority need. Council tenancies are increasingly held by females, either older women living alone or lone mothers (Ungerson & Kember, 1997). Yet, strategic direction in housing has been driven by stock allocation and management priorities rather than community considerations of local needs, including the requirement for appropriate housing for homeless women with mental health problems (Franklin, 1998).
Employment and gender
Paid employment is protective of good mental health, providing a role, status and financial support. Female workforce participation has increased since the 1970s, but on a part-time basis, and particularly for White women. The occupations in which women participate are mainly segregated into particular sectors: education, health and financial and hotel services. Child care responsibilities and the financial consequences of finding a carer may prohibit women from sustaining full-time work, and nearly half of working women are part-time. Working part-time allows women more contact with their children than full-time work, but it affects the chance of promotion, earnings, employment protection, eligibility for benefits and sick pay and access to occupational pensions (Gosling et al, 1997; Ungerson & Kember, 1997).
Lesbian and bisexual women with a mental illness may be a largely ignored or invisible group, within both the lesbian community and mental health services. They are often dependent on support from families and services that make assumptions about women being heterosexual and neglect this group's special needs. In addition to dealing with the stigma of mental illness, they may also face the stigma attached to their sexuality (Hellman, 1996).
Although the majority of homeless people are male, up to one in six is female. Younger women are more likely to stay with friends, while older women may be in direct-access hostels. Homeless women have stronger social support networks than homeless men, but higher levels of mental illness despite lower levels of substance misuse (Marshall & Reed, 1992). Local authorities are obliged to rehouse vulnerable people in priority need, for example, pregnant women, households with children, the elderly and those with mental health problems, but there is no obligation to rehouse women because of domestic violence.
Women from ethnic minorities
Our understanding of mental health problems follows a Western conceptualisation of mental illness, and women from ethnic minorities may face additional barriers such as language difference in engaging with health care professionals. Factors such as this may partly explain the underrecognition of mental illness in these groups, for example, in Indian women in the UK (Jacob et al, 1998). Women from refugee groups are particularly likely to have experienced multiple losses and may be at risk of depression and PTSD.
The proportion of women living alone increases in old age. Older women are more likely to have smaller incomes and be reliant on the informal care of female relatives, and may face or have already faced loss, including bereavement. Livingston & Blanchard (1996) discuss the tendency of the current generation of older adults to describe themselves as well, despite answering positively to questions about a number of distressing symptoms. They suggest that this is a cohort effect and later generations of older people may articulate their needs differently. Professionals tend to regard the current generation of older women as having generic ‘old lady’ status rather than identifying them with their previous occupations, as they would older men, so ignoring some of the socio-economic complexities of their lives.
Needs assessment in women patients
Recent government mental health policy has emphasised the importance of assessing individual needs.
How can needs be assessed?
Over the past decade, health care needs assessment has assumed a larger role in the process of health care planning, prompted by the rapid growth in health service spending, and a concurrent concern about effectiveness and appropriateness in targeting resources (Stevens & Raftery, 1994). Recognition of the importance of a needs-led approach has gained particular prominence in the development of community mental health services. This recognition has led to the development of a variety of methods for assessing needs.
Concepts of need
Approaches to needs assessment have varied according to concepts of need. For example, health economists have looked at needs from the perspective of supply and demand, while public health physicians and epidemiologists have aimed at investigating unmet need. A more recent definition of health care need has evolved, which is the “population's ability to benefit from healthcare” (Stevens & Raftery, 1994). Here, it can be seen that ‘need’ will often differ from supply or demand. For example, there may be a demand for a treatment that has been shown to be ineffective. On the other hand, effective interventions may exist for which there is no demand, because individuals are not aware of their problem, not aware that an effective intervention is available or are unable to communicate their wishes. Supply, on the other hand, often reflects historical patterns together with public and political pressures rather than assessed population needs.
The ‘ability to benefit’ definition of need does not have to apply narrowly to ‘treatment’. We can also apply it to prevention, diagnosis, continuing care, rehabilitation and effective reassurance. Although this paper focuses on the needs of patients, the same approach to need can be applied to the support and relief of carers. The scope of attention in needs assessment can also change. For example, Stevens & Gabbay (1991) have described a shift of attention from the focus in the 1970s and 1980s on issues of relative deprivation and resource allocation at a population level, to the development of finer grained assessments of the needs of specific groups.
Needs assessment tools
Stevens & Raftery's (1994) thorough two-volume resource book on health care needs assessment contains several chapters on mental health care, including adult and child mental health, dementia, drug and alcohol misuse and learning difficulties. However, the references to women's needs are mainly confined to information on gender differences in the prevalence of psychiatric disorders and to needs associated with child care. There is now a growing range of well-developed standardised assessments of needs for mental health care. These include: the Medical Research Council's detailed MRC Needs for Care Assessment (Brewin et al, 1987) and its modified version, developed by Marshall et al (1995); the Cardinal Needs Schedule, with an accompanying instrument for assessing the needs of relatives; and the Camberwell Assessment of Need (Phelan et al, 1995), which was developed as a comprehensive needs assessment tool for people with severe mental illness. Other needs assessment instruments have been developed for other patient groups, such as those with dementia. These are broad-ranging tools, and while they are not designed specifically to assess the needs of women, they cover many areas in which identification of gender differences in need could be helpful.
Needs assessment and mental health policy
1989 Caring for People: Community Care in the Next Decade and Beyond(Department of Health, 1989) – assessment of need is a “cornerstone of high-quality care”
1991 Introduction of the Care Programme Approach, to include “systematic arrangements for assessing the health and social needs of people accepted into specialist mental health services”
1998 Modernising Mental Health Services: Safe, Sound and Supportive(Department of Health, 1998) – “The needs of people in inner cities are very different from the needs of people in the countryside, and assessment of population need is a sound basis for planning and commissioning services … Individual needs assessment is the next step”
1999 Saving Lives: Our Healthier Nation(Department of Health, 1999a) identifies effective interventions, for example, teaching parenting skills, social support groups for young isolated mothers, rapid treatment for mothers with depression, help at school for children whose parents are divorcing and practical information for those caring for people with dementia
1999 The National Service Framework for Mental Health(Department of Health, 1999b) identifies programmes for individuals at risk and vulnerable groups
1999 Effective Care Co-ordination in Mental Health Services: Modernising the CPA(Department of Healthet al, 1999) considers diversity through attention to service users' culture, ethnicity, gender and sexuality
2000 Secure Futures for Women: Making a Difference(Department of Health, 2000a) discusses the needs of women across the spectrum of mental health care, but with an emphasis on the needs of women in secure services
2000Safety, Privacy and Dignity in Mental Health Units(NHS Executive, 2000b) considers the needs of and risks to men and women and provides guidance on mixed-gender accommodation
Using such tools, it should be possible to investigate the mental health care needs of women in a variety of settings. However, this may pose challenges in the application, and where necessary adaptation and validation, of needs assessment instruments to address women's needs. In Abel et al's (1996) book on planning community mental health services for women, several authors have indicated the lack of adequate needs assessment as a basis for service planning. For example, Subotsky noted that youth provision has previously been targeted more at those young people presenting as ‘troublesome’, among whom boys have predominated. With increasing knowledge of the epidemiology of childhood and adolescent psychiatric disorder and the use of standardised screening instruments, the needs of older girls with emotional disorders are becoming more apparent. Social changes in women's roles will affect all age groups and will necessitate revisiting of earlier needs assessments.
The recent growth in the development of needs assessment measures in mental health care provides a good basis for further examination of the extent to which these measures work for women, and if necessary, for the development of specific measures to address women's needs. These can be epidemiologically based, with attention to gender differences in different settings.
In addition to the development of needs assessment measures, the increasing use of routine performance monitoring and clinical audit, together with increasing participation of service users in service evaluation, is likely to give us more information on the needs of women with mental health problems. For example, a recent initiative to reduce the waiting list for an in-patient unit for the treatment of alcohol problems in part of our trust led to careful monitoring of referrals and admissions. Although not the focus of this exercise, we discovered that women referred to the unit were less likely than men to take up a place, even though they were in contact with community services, were apparently appropriately referred and were offered admission. This sort of information can lead to further investigation of the needs of women patients and the potential barriers for women in making use of services that are available for them.
Multiple choice questions
Considering gender effects in the prevalence of mental illness:
there is no significant difference in presentation and course of mental illness in men and women
there are some women who become depressed only after childbirth, in relaiton to motherhood and its demands
the suicide rate is significantly reduced in the year after childbirth
alcohol and drug misuse and antisocial personality disorder are equally common in men and women.
Reviewing gender differences within the social context:
in the 1970s and 1980s there was a four-fold increase in the number of divorced and separated women with children, and a ten-fold increase in the number of lone mothers
mothers who have themselves had poor experiences of parenting may be more likely to receive practical help with childcare and support
a high proportion of women attending A & E departments report a history of domestic violence
victims of completed rape are at increased risk of suicide atrempts and of having a depressive illness.
In assessing needs:
one definition of need is the “population's ability to benefit from healthcare”
the ability to benefit applies only to treatment
the focus of needs assessment has shifted from issues of resource allocation at the population level to assessing the needs of more specific groups
there are good needs assessment tools for women patients in different settings.
- © The Royal College of Psychiatrists 2001