A Proposal for a Training Programme for Community Mental Health Nurses in the Assessment of Suicide Risk

Dissertation

B. Sc. Community Health Studies 1998/99 University of Wales College of Medicine - School of Nursing Studies

Steve Wood

Statement of Originality

I declare that all the work in this study is my own except where specific references have been made. I further declare that this work has not been submitted towards any other degree and is not being currently submitted in candidature for any other degree

Signed ......

Date ......

Abstract

This is a proposal for an educational innovation, a Training Module for Community Mental Health Nurses (CMHNs) in the assessment of suicide risk. Against a background of rising rates of suicide and parasuicide among people in younger age groups, the dissertation discusses strategies for achieving government targets for reducing suicide. Specific groups of people are identified, with whom CMHNs routinely have contact, and who are at high-risk of suicide and parasuicide. Assessment of suicide risk is discussed, and, in particular, aspects of it which can be regarded as requiring additions to the skills and knowledge normally employed by CMHNs. Proposals are set out for a specific training module in the assessment of suicide risk by CMHNs, taking into account relevant learning theory. Finally, possible development of the training module is discussed.

Introduction

Summary

This dissertation proposes an educational innovation, set in the context of two recent public health Green Papers (Department of Health 1998, Welsh Office 1998) and the Report of a National Confidential Enquiry (Department of Health/Royal College of Psychiatrists 1997). The Green Papers set targets for the reduction of mortality from suicide and undetermined death. The Enquiry Report identified better recognition of the individuals most at risk as a key objective for mental health services.

The proposal is for a Training Module for Community Mental Health Nurses (CMHNs) in the Assessment of Suicide Risk. The dissertation aims, firstly, to make a persuasive case for this based on evidence of need. The second aim is to describe a means for doing so based on evidence of best practice.

The dissertation consists of four main sections. The first looks at trends in the incidence of suicide and parasuicide, and describes groups within the community which are at increased risk. Attitudes and policy in this area are examined, and some of the difficulties in achieving the targets set by policy-makers discussed.

The second section starts by discussing the close association between suicidal behaviour and mental disorder. It goes on to focus upon particular groups of people, shown by the literature to be at high risk, with whom CMHNs routinely have contact in the performance of their job.

Section three examines the process of assessment of suicide risk, by reference to some of the extensive literature. It discusses the uses and limitations of predictive scales, some commonly held misconceptions about suicide, and the hazards which are present in the process. It concludes by outlining recommendations for good practice.

Based on the previous sections, the concluding chapter sets out the rationale, aims and expected learning outcomes of the Training Module. The proposed structure, training methods and evaluation are outlined, with reference to learning theory. Finally, development and extension of the module are discussed.

Literature Search Strategy

The literature on the subject of suicide is voluminous. The following search terms were used: incidence of suicide; community psychiatric and/or nurse and/or nursing in combination with prevention of suicide and suicide risk assessment. The following databases were searched, to identify English language papers: CINAHL, AMED, ASSIA, EMBASE CD:Psychiatry 1988-95, EMBASE CD: Psychiatry 1996-98 and Clinpsyc 1988-98.

The search term incidence of suicide generated several hundred references. This was not the main focus of the dissertation, and is not an area of controversy: therefore, papers were selected to give an overview of suicide trends during the present century, particularly over the past three decades.

Wherever possible, references generated by the other search terms were obtained. Further relevant references, cited in those articles but not generated by the original searches, were obtained.

In addition, the website of the Samaritans was accessed.

Chapter 1 - Suicide & Parasuicide

Definitions; Incidence and Trends; Risk Factors; Attitudes and Policy; Prevention and Reduction

This section is concerned with the context of the relationship between CMHNs and suicide risk assessment. It examines the phenomena of suicide and parasuicide. It describes how, against a background of decline in overall suicide rates, suicide and parasuicide among younger age groups, both male and female, have risen. Major risk factors and precipitants are outlined. Changing attitudes are discussed and the policy background explained, including the recent move towards setting targets for suicide reduction. With these in mind, some of the difficulties in prevention and reduction are set out.

Definitions

The most frequently quoted definition of suicide was penned in 1897 by Durkheim (translated in Spaulding & Simpson 1952): “The termination of an individual’s life resulting directly or indirectly from a positive or negative act of the victim himself which he knows will produce this fatal result”. It has been described as “the most important consequence of psychiatric disorder” (Appleby 1992, p 749) and a “major preventable cause of death” (Murphy 1983, p 343).

Various terms have been used to describe the phenomenon of non-fatal self-injurious behaviour. The word “parasuicide” was coined by Kreitman et al (1969) as an alternative to “attempted suicide”, a term criticised “for the excellent reason that the great majority of patients so designated are not in fact attempting suicide”. The word attempted to describe a form of behaviour which simulated suicide but could not “be construed in any simple sense as orientated primarily towards death” (p 746).

Similarly, “deliberate self-harm”, was first used by Morgan (1979) to include “self-poisoning” and “self-injury”, both of which imply that physical harm always occurs, which is not so.

Hawton & Catalan (1987) consider suicide and parasuicide to be separate phenomena, with an area of overlap. They state that rates of each changed independently of each other during the 1960s, 1970s and 1980s: also, that parasuicides and completed suicides differ in demographic characteristics and methods used.

Whether or not this is so is largely academic from the point of view of CMHNs in their daily practice, given the absence of definite distinguishing characteristics, other than outcome. Moreover, Farmer (1988) expressed the view that parasuicide was relevant to the study of suicide, saying that to treat it separately was akin to studying accidents only by examining fatalities.

Incidence & Trends

Different criteria used to calculate suicide rates leads to large international variations. Generally speaking, intranational is more reliable than international analysis (Morgan 1979).

Suicide is rare. In most countries it accounts for between half and one per cent of all deaths. Generally speaking, the rate increases with age Its impact upon mortality is greater among the young (Farmer 1988). In Britain, it is currently the second most frequent cause of death in the 15-34 age group. It accounts for two per cent of male, and one per cent of female, deaths before age 65 (Gunnell 1994). In the United Kingdom in 1996, there were 5905 suicides (Samaritans 1999).

In Britain between 1901 and 1970, suicide declined in older males and rose in women of all age groups (Sainsbury 1973, Morgan 1979). Between 1955 and 1965, an increase in suicide by males under age 25 began. The late 1960s saw a substantial fall in all age groups except the youngest. Since 1970, there has been a steady rise in the rate for males aged between 25 and 34, and for females aged 15 to 24 years, and 35 to 44 years. In a study of suicides over 20 months in Bristol (Vassilas & Gunnell 1994), the percentage of suicides by people aged under 35 years was much higher than earlier studies, e.g. Seager & Flood (1965), Ovenstone (1973).

Latest estimates of the incidence of parasuicide are 140,000 cases annually in England and Wales, an increase of 50 per cent since 1990 (Samaritans 1999). In a review of 23 studies, covering 6 countries, Weissman (1974) reported a sharp increase in parasuicide during the 1960s in the USA and other western countries. Wexler et al (1978) noted that the rise in the USA was levelling off in the mid-seventies. Most countries studied, the UK, the USA, Australia, Canada, Sweden, Israel, Singapore, and the West Indies had rates above 100 per 100,000 population.

In the late 1970s and early 1980s, there was a decrease in rates, particularly in older teenage girls (Platt et al 1988). In the late 1980s, rates started to rise again, especially amongst older adolescent females (Hawton & Fagg 1992). In 1987, Hawton and Catalan reported that deliberate self-poisoning was the commonest reason overall for admission to acute medical wards, and the second most common reason for men.

The rising trend has continued into the present decade. Hawton et al (1997) reported an increase in deliberate self-harm in Oxford of 62 per cent for males and 42 per cent for females in the period 1985 to 1995. The greatest rate of increase was amongst males in the 15 to 24 age group. Changes in rates of deliberate self-harm correlated with changes in national suicide rates for males and females in this age group. Repetition rates for both men and women also increased during the period under study. Rates of deliberate self-harm in the United Kingdom are amongst the highest in Europe.

Risk Factors

There are many risk factors. Bagley et al (1973) commented that interrelationships between them were more complex than they originally hypothesized. Societies with high rates of suicide are those subject to economic instability, deprivation, breakdown of traditional family structures, increasing criminal behaviour and interpersonal violence, secularisation, and increasing substance misuse (Diekstra 1991).

In England and Wales, male suicides outnumber female by about 3:1. Increased rates are seen in the highest and lowest social classes. The rate amongst unemployed people is between 2 and 3 times that of the general population: the relationship involved is complicated and has not been shown to be direct cause and effect (e.g. Banks & Jackson 1982, Platt & Kreitman 1984, Smith 1985, Jackson & Warr 1984). Rates are highest amongst single, widowed and divorced men (e.g. Barraclough et al 1974, Charlton et al 1994). Certain occupational groups are associated with high rates: typically, these are in the agricultural and caring professions (Charlton et al 1994).

Parasuicide is rare before age twelve, becomes much commoner in young adult groups, declines substantially in middle age and is rare after age 60 (Hawton & Catalan 1987). Females outnumber males in all international studies (Wexler et al 1978), mostly by about 2:1 (Weissman 1974). The greatest preponderance of females is among adolescents, for example 4:1 in a study in Oxford between 1974 and 1979 (Hawton & Catalan 1987). There is an excess of single, separated and divorced people (e.g. Weissman 1974, Morgan et al 1975a). A high degree of social mobility is seen, e.g. 40 per cent having changed address within the past year (Morgan et al 1975b). Parasuicides have been characterised as isolated, rootless and drifting (Gibbons et al 1978). An excess has been reported in the lower social classes (Weissman 1974, Paykel et al 1975); in areas of houses in multiple occupation, overcrowding and lack of exclusive domestic amenities (Morgan et al 1975b); and in areas with a high proportion of foreign-born residents (Morgan et al 1975b).

Morgan et al (1975a) reported that 36 per cent of male, and 26 per cent of female, suicide attempters were unemployed. Holding et al (1977), in a study in Edinburgh between 1968 and 1974, found no clear association between the unemployment rate in the general population and parasuicide. However, associations between parasuicide and unemployment have been reported (Buglass & Duffy 1978), and also between parasuicide and indicators of poverty (Bagley et al 1973). A study of males in Edinburgh between 1968 and 1982 (Platt & Kreitman 1984), reported that parasuicide was ten times more frequent amongst unemployed men and that this risk was doubled in those out of work for more than a year.

There is a very strong association between psychiatric disorder and suicide. Studies of completed suicides have variously reported the frequency of a previous history of psychiatric illness: e.g. 33 per cent (Seager and Flood 1965); almost 60 per cent (Bunch et al 1971); 50 per cent (Barraclough et al 1974, Vassilas and Gunnell 1994); more than 50 per cent (King and Barraclough 1990).

A high proportion of completed suicides have made previous attempts. This has been reported as, e.g. 16 per cent (Seager & Flood 1965); 33 per cent, more than half of those within the preceding year (Barraclough et al 1974); 47 per cent (Ovenstone & Kreitman 1974); and 39 per cent (Vassilas & Gunnell 1994).

Weissman (1974) reported that 1 per cent of failed attempters complete suicide within the next twelve months, rendering them about 100 times more likely to commit suicide than the general population. In a comparison of 64 suicides with 128 living depressed controls, Wexler et al (1978) noted that previous attempts were seven times commoner among the suicides. Hawton and Catalan (1987) stated that more than 2 per cent of attempters go on to commit suicide in the next 2 to 5 years. A follow-up of 974 patients who had attempted suicide found that, ten years later, 103 had succeeded (Nordentoft et al 1993).

Alcoholism has been consistently reported as the principal diagnosis in one in every six or seven suicides. For example, 13 per cent (Sainsbury 1973); 15 per cent (Barraclough et al 1974), 18 per cent (Dennehy et al 1996). The increased risk in people with alcohol dependency has been reported as 2.3 times for women and 3.2 times for men (Charlton et al 1994). There is a marked association between alcohol problems and parasuicide, particularly among men. Morgan et al (1975a) reported alcoholism in 18 per cent of 338 people admitted to hospital after attempted suicide. The links with alcohol abuse are long-term effects on social adjustment, and the short-term disinhibiting effects of alcohol when suicidal urges are felt. Weissman (1974), in a systematic review of 23 studies, noted alcohol intake immediately before an attempt as characteristic. In Peterson et al’s (1985) study of 30 attempters, half had used alcohol or drugs immediately prior. In a 1984 study in Oxford, 25 per cent had consumed alcohol shortly before attempted suicide (Hawton & Catalan 1987).

Increased risk amongst heroin addicts has been reported as 20 times (Charlton et al 1994). Prisoners are another high risk group, accounting for two per cent of suicides in the 15-44 age group (Charlton et al 1994). There were 300 prison suicides in England and Wales between 1972 and 1987: the increase in the rate amongst prisoners greatly exceeded the increase in the prison population (Dooley 1990). Excess prison suicides are by young, male remand prisoners. Previous history of psychiatric treatment and self-injury are frequently reported.

Increased incidence of poor physical health among suicide attempters has been reported (Hawton & Catalan 1987). Epilepsy was noted to be between five and seven times likelier in parasuicides than the general population, especially among men (Hawton et al 1980).

Life events associated with suicide and parasuicide have been reported to be recent bereavement or loss (e.g. Bunch et al 1971, Dorpat & Ripley 1960, Levi et al 1966), bankruptcy, impending court cases, and serious physical illness (Seager & Flood 1965).

Weissman (1974) reported interpersonal disorganisation and breakdown of personal resources to be characteristics of parasuicides and found an association with recent, serious interpersonal conflict. Morgan et al (1975a) reported this to be the case in 51 per cent of cases, mostly with a key individual, for example a spouse or co-habitee. A study of 159 subjects reported significantly more recent undesirable life events than matched groups of depressed people and general population controls (Paykel et al 1975). This was supported by Bancroft et al (1977), who noted that events involving a key relationship were much commoner than other kinds, especially involving upsetting or rejecting behaviour. The most important event was a quarrel, particularly in the 48 hours prior to the attempt.

Recent studies continue to show relationship problems are the commonest precipitants for episodes of parasuicide, concerning a partner in the case of males, and family members other than a partner in females. Other frequently reported precipitants were problems with employment or studies; alcohol or drugs; and finances (Hawton et al 1997).

The following chapter considers in more detail some of the risk factors mentioned above, with particular reference to their importance for CMHNs.

Attitudes and Policy

Current policy reflects the prevailing view that suicide is undesirable and ought to be prevented. This has not always been so. In an historical review, Morgan (1979) described how there has always been evidence of a debate about its nature, whether it is essentially sinful or can be justifiable. In Ancient Greece, the Senate could permit suicide by drinking hemlock if a case had been made. In Ancient Rome, suicide was not an offence unless it was carried out by a soldier, a slave, or someone accused of a crime. From about the fourth century, when St Augustine declared it “a greater sin than any one might avoid by committing it”, a repressive attitude persisted until the present century.

In mediaeval times, orthodox Christianity held that the soul of a suicide was condemned to the eternity of hell. In many European countries, until well into the nineteenth century, the goods of a suicide were forfeited. Since the Renaissance, some writers have been apologists for suicide, e.g. Erasmus, Thomas More and John Donne. Sym, in 1637, gave the first advice about recognising suicide risk and preventing suicide. Signs of suicidal intent were said to be abnormal behaviour, talking to oneself and the verbal expression of threats and warnings. It was suggested that those at risk should avoid solitude, darkness, bridges and precipices, and take care with knives.

Since the Suicide Act 1961, attempted suicide has been regarded as entirely a medical and social problem, as opposed to a criminal offence. In 1968, the Hill Report (Central & Scottish Health Service Councils 1968), made recommendations regarding the management of attempted suicide. These included establishing poisoning treatment centres in district general hospitals; psychiatric emergency cover 7 days a week; all cases being referred for assessment; psychiatric history and examination taken for all; social work help available in all cases; and continuity of care after discharge. The necessity for psychiatrists to carry out all assessments was questioned (Crammer 1969), and it was demonstrated (Lawson & Mitchell 1972) that an acute medical unit without special facilities could meet the Hill Report recommendations. In practice, psychiatric referral was not made in all cases.