Chapter 3
Who is at risk of depression?
One advantage of the more consistent application of diagnostic criteria for identifying individuals with a depressive disorder is that it allows national and international comparisons to be undertaken. Large-scale studies make it possible to estimate the overall prevalence of depression, and repeating the surveys allows detection of any changes in these rates over time. Comparisons can be made between the distribution of depression cases by country, culture, economic and social status, and other demographic features such as age, gender, marital status, etc. Differences between any of these subgroups can offer important insights into who is at risk of depression, at what time point an episode is most likely to occur, and also the development of theories about what factors reduce the risk of or protect against such experiences.
We explore the epidemiology of depression (which refers to the distribution and determinants of depression-related states), give examples of the presentation of depression across the lifespan, and discuss some gender-related issues. Lastly, we highlight some of the current thinking on the prevention of suicide.
Epidemiology
The World Health Organization (WHO) has estimated that over 5 per cent of the global population will be depressed during any one year and that about 15 per cent of the population will experience a depression at some point during their lifetime. On average an episode will last between four and eight months, but recurrences are common and about 50 per cent of depression cases will have at least one further episode of depression within five years. Lamentably, the WHO also reports that only 25 per cent of those who experience a depressive episode will have access to effective treatments. However, behind these headline figures there is considerable variation in the estimated rates—so here are a few examples that have been chosen simply to illustrate some of these differences. We emphasize that the list of issues we discuss is not exhaustive, but the topics have been selected to show how researchers use these data to begin to develop theories about why certain sub-populations are more or less likely to become depressed.
Geography
The prevalence of depression is not uniform across countries or continents. For example, rates are reported to be especially high in France and America. The reasons for this are unclear; there are similarly high rates in India, whilst some of the lowest rates are reported in Taiwan and China. There is some suggestion that it is not geography but Gross Domestic Product (GDP) that may in part explain these different rates—with depression being more common in high-income countries (about 15 per cent) compared to low- to middle-income countries (just over 10 per cent).
There has long been a suggestion (first put forward by researchers in schizophrenia) that agrarian societies may be less stressful places to live than industrialized or urban environments and also that such communities may be more tolerant and supportive of individuals with depression. It is unclear if this can explain the geographical variations in depression rates, but the data from countries such as India are interesting given that it is a continent that is undergoing significant social and economic changes. It is proposed that transitional regions may show greater instability because the conflict in values between metropolitan and more rural settings potentially increases stress levels in these areas. The theory suggests that individuals who are vulnerable to depression are therefore more likely to experience an illness episode than similar individuals living in more stable regions. In contrast it is argued that the lower rates of depression reported in some countries (such as China) could be explained by the fact that some individuals or social groups are still less likely to recognize, acknowledge, or seek help for psychological problems.
Culture and ethnicity
Depression and its symptoms may be expressed differently in different ethnic groups and cultures. A simple example of this phenomenon comes from one of the recent large-scale community studies undertaken in America that was published in 2012. The one-year prevalence of depression was virtually the same in Hispanic and White Americans (about 7 per cent), slightly lower in Black Americans (just over 6 per cent), about 3 per cent in Asian Americans, but about 10 per cent in Alaskan natives. It is possible that the reported rates are influenced by the nature of how depression is perceived by the individual. It is known that in some cultures or ethnic groups more attention is given to physical experiences (such as low energy, poor appetite, and disturbed sleep) rather than to psychological or emotional symptoms of depression. For example, individuals from Asian countries or cultures are more likely to report physical symptoms. So, it is possible that publications that indicate low rates of depression are underestimating the actual rates. Alternatively, it could be that the different rates are not due to differences in reporting of symptoms, but that there may be specific risk or protective factors operating within these different ethnic groups or cultures that modify the level of risk for developing depression in different sub-populations in the USA.
Socio-economics
A 2010 survey comparing rates of depression in Germany, America, and England found that depression was most prevalent in the poorest sub-sample of respondents (18–27 per cent) and lowest in the wealthiest sub-sample (4–10 per cent). Other research indicates that rates of depression are about three times higher in those who are unemployed as compared to those who are employed. These data are often viewed as controversial, mainly because they are interpreted (used and misused) in different ways by different political groups. However, it must be recognized that evidence of increased prevalence rates does not in itself explain the direction of causality, that is, we cannot assume that unemployment increases the likelihood of depression, as it is possible that the depression came first and reduced a person’s ability to gain or maintain employment, with knock-on effects for income levels and quality of life. Indeed, in this example, the relationship may well be bidirectional, with unemployment increasing the risk of depression and depression increasing the risk that someone will be unemployed.
Chronology
In international studies the average age of onset of the first episode of depression is the mid-to-late twenties; and it has been reported that the first experience occurs about two years earlier in lower- as compared to higher-income countries. In large-scale community studies undertaken in America, the one-year rate of depression was higher in 18–25-year-olds (around one in ten) than in any other age group. Worldwide, about 40 per cent of people report the initial episode of depression occurred before the age of 20 years, about 50 per cent report it began between 20 and 50 years, whilst only 10 per cent state their first experience of depression occurred after the age of 50 years.
Interestingly, there is evidence that the rates of depression and age of onset have changed over the last fifty to sixty years and that the risk of experiencing at least one episode of depression has increased and the age of onset of the first episode has decreased in individuals born after the Second World War. Suggested reasons for these temporal changes vary from the notion that the increase in reported rates of depression offers evidence of medicalization, that is, that normal sadness is being misdiagnosed as illness. Other suggested explanations are that the increase in rates of depression is an artefact of increased access to health services for all members of society (i.e. depression rates are unchanged, but detection has increased) or that more people are prepared to seek help.
If medicalization or changes in treatment seeking do not explain the observed changes in rates of depression and age of onset of the first episode, it is interesting to consider other reasons. The relatively short time frame during which this increase has occurred means it is unlikely to be explained by genetics as changes in our genetic makeup only become apparent after many hundreds of years. However, social and environmental changes can have an impact within a few decades on our health and well-being. For example, research suggests that increased exposure to drugs and alcohol in the post-war years may partly explain the increase in depression.
Gender
Rates of depression in women are consistently reported to be twice those reported in men. As this gender gap is found in surveys of both untreated as well as treated populations, it cannot be solely attributable to a greater tendency to recognize, report, or seek treatment for distress in women. Other explanations have been put forward, ranging from the influence of hormones to social role differences, and will be discussed in Chapter 4.
Marital status
Cross-culturally, the loss of a partner, whether by death, divorce, or separation, is associated with increased rates of depression. Married men have the lowest rates of depression and separated or divorced men have higher rates. There is a less clear association in women. Many explanations for these findings have been proposed, but the answer is not straightforward. For example, it is uncertain whether having depression causes a marriage to fail or if the stress of divorce or separation (or the reason for it) is the cause of depression. Alternatively, another independent factor such as abnormal personality could increase the likelihood someone will become depressed and also interfere with their ability to maintain a long-term relationship.
Depression across the lifespan
In the remainder of this chapter we explore depression in childhood and adolescence, depression in women of childbearing age, and depression in men and co-occurring with physical health problems. Then we highlight issues related to suicide (which is commonest in younger and older adults).
Childhood
For many years, depression was regarded as a disorder of middle-aged and older adults, and children and adolescents were thought to be immune to such experiences. Furthermore, the few dissenting voices who suggested that depressive disorders could have an onset in childhood had little evidence to draw on as participation in research was usually restricted to individuals aged over 18 years. From about 1975 onwards, a number of mental health research institutions began to question the perceived wisdom that childhood depression did not exist and several sophisticated long-term follow-up studies were commenced in which children and adolescents were assessed on several occasions over a number of years to examine how many children became depressed and how many of this group experienced repeated episodes of depression, developed manic depression or other mental health problems, and how many had a single depressive episode without any further mental health problems. In reality, the most important findings from these studies extend beyond simple number crunching, as they offer some important insights into risk and protective factors and the differences in the patterns of depression seen in boys and girls before and after puberty.
In children under the age of 11 years, depression is relatively uncommon. In these pre-pubertal children, there is no evidence of the female predominance that is seen consistently in all other age groups; indeed some studies suggest that the prevalence of depression in young boys may actually be higher than in girls. Interestingly, depression in many children does not occur in isolation and the symptoms are often mixed up with anxiety or irritability. Furthermore, depression is not usually the first problem that the child experiences. In about four out of five young children, the development of depression is a complication of other difficulties such as autism or disruptive behaviour problems. Some researchers speculate that the symptoms of depression observed in these children (such as low energy and altered sleep) may represent an ‘exhaustion syndrome’ that arises as a consequence of the high level of stress associated with their other problems. This idea is important as it overlaps with theories about the role of an overactive stress hormone system in causing depressive symptoms (see Chapter 4). Another noteworthy finding from these large-scale studies is that children who have a family history of depression (e.g. a parent or grandparent who has been treated for depression) are four times more likely than other children to experience an episode of depression early in their life, and if they have one episode, they are more likely than other children to have further episodes.
Given the scepticism in some quarters about whether depression in adults is a ‘manufactured’ condition, the reports of depression in children aged 5–11 years have opened a whole new can of worms, not least because of the implications for treatment interventions. Many clinicians are understandably reticent about prescribing medications developed for adult disorders to younger people. Studies suggest that talking therapies such as cognitive behaviour therapy and family approaches can be useful, but recent efforts have also turned to the notion that it is important to try to reduce the likelihood of developing depression by increasing resilience in larger groups of young children who may be at risk. For example, this has led to the exploration of any benefits of including mental health promotion within the school curriculum and the introduction of social-emotional learning (SEL) classes. More specific depression prevention strategies have included projects that offer training in mindfulness for children.
Adolescence
Anyone who has spent more than a few days in the company of an adolescent will be aware that mood states, sleep patterns, and self-esteem are highly variable and transient, but also that intense distress can be quite common. As such, there is a considerable challenge in being certain about when normal adolescent unhappiness evolves into a clinical depressive episode that warrants treatment interventions. That said, some of the most recent research indicates that the prevalence of depression in adolescents is the same as in older adults.
One of the important insights from depression research in adolescents is that it is not chronological age but puberty that seems to herald a sharp surge in reported rates of illness. This suggests that hormonal changes may be important, a hypothesis supported by findings that rates of depression in the post-pubertal period are twice as common in young women as young men. As noted in younger children, those adolescents with a family history of either depression or manic depression have a greater risk of developing depression in early adulthood than those without any family history and a family history is more often found in those who develop recurrent episodes of mood problems.
There are many life events that are experienced by young people that are developmentally normal, but that may still trigger episodes of depression. Issues with peer groups, relationship breakdowns, coping with leaving home, and exposure to drugs and alcohol can all precipitate the onset of episodes of depression especially in individuals who may be more vulnerable to depression for other reasons (such as a family history of mood disorders). Also academic performance and economic issues can play a part in this age group. For example, young people who are not in employment, education, or training (so-called NEETs) report depression rates that are three to five times higher than their non-NEET peers. Whilst it is hard to disentangle cause and effect, this finding serves to demonstrate that any interventions for depression in adolescents and young adults cannot be limited to treating depressive symptoms and may well need to include help with re-engagement with social and academic networks.
Unsurprisingly, adolescents frequently show ambivalence about taking medications for depression. Furthermore, young men do not always find it easy to engage with other available treatment interventions such as talking therapies. Some of the solutions proposed to resolve this dilemma include the use of activity- and behaviour-orientated groups to help tackle the symptoms of depression and the exploration of how to use electronic media such as internet applications or web-based programmes. In countries such as Australia research is being undertaken where these options are offered to all school pupils in a particular academic year (e.g. those taking final school examinations that determine their prospects for entry into higher education). Pupils in these school years are targeted on the basis that the rates of depression can be predicted to rise in the face of such stressors and that prevention may be better than cure.
Many young people who do go through a period of depression will find that such psychological problems are confined to adolescence. However, for others it heralds the start of a condition that can affect them for many years. Trying to identify young people who are most likely to develop recurrent mood episodes is a major research priority. Furthermore, it is important to try to differentiate between individuals who may experience repeated episodes of depression and those who may develop manic as well as depressive episodes. There are only a few clues so far. For example, we now know that 70 per cent of individuals who go on to develop bipolar disorder (or manic depression) in early adulthood report that they had a depressive episode in adolescence. Also, this episode often occurs at a slightly earlier age compared to those who get recurrent depression. However, identifying those young adults with a history of depression who may also be at risk of mania is a considerable challenge, as behaviours that can be part of a manic presentation such as risk taking, disinhibited behaviour, staying up all night, and being the life and soul of a party are not necessarily symptoms of illness in late adolescence.
Currently, having a family history of bipolar disorder is one of the few factors that may identify which young people are more likely to experience mania in the future. The limited ability to predict future bipolar disorder with any certainty is a significant barrier to effective treatment. Several surveys of individuals with bipolar disorder identify that the delay in identifying the problem and offering the most appropriate interventions is one of the biggest issues for patients and their families.
A further issue in trying to identify which individuals will ultimately develop bipolar disorder is that even someone with an above average risk is less likely rather than more likely to ever experience an episode of mania. It is currently estimated that less than one in three individuals with multiple risk factors will develop a full-blown bipolar disorder. As such, it is not rational to start prescribing treatments that are used routinely for older adults with an established illness to individuals at risk of, but who do not have a diagnosis of, bipolar disorder. Some researchers have focused on developing interventions with a high benefit to risk ratio. These include non-medical approaches such as lifestyle management and psycho-education programmes that can help a young person to manage any fledgling symptoms or cope with social problems without the risk of side effects or adverse effects that might be seen with some medications. However, there is not yet sufficient evidence to support the introduction of these strategies in day-to-day clinical practice.
Depression in women of child-bearing age
Rather than considering all manifestations of depression in women, we discuss two depressive disorders associated with child bearing, namely post-natal depression and puerperal psychosis.
The birth of a baby is often a reason for celebration and a post-natal depression associated with such an event is frequently regarded as inexplicable by those outside the immediate family circle. It is easy for people to accept that a new mother may feel emotional or weepy in the days immediately after delivery of a baby, when hormone levels come crashing down, physical exhaustion kicks in, or both parents feel overwhelmed by the responsibility of caring for a new baby. However, these transient ‘baby blues’ are not the same as a more intense and persistent depressive episode which needs to be viewed as a very serious problem that requires early intervention. Any treatment offered must also address directly the feelings of guilt that are expressed by the mother about becoming depressed.
In most ways, the signs and symptoms of post-natal depression reflect those seen in depressions that occur at other times, but what sets post-natal depression apart is the potential impact on the baby. Not only does the depression impair the mother’s self-care and her quality of life, but it may affect the day-to-day care of the baby. Importantly, it can complicate plans for breastfeeding as some antidepressant medications are secreted in breast milk and would be passed to the baby via this route. Depression at this time can affect the bonding process between a mother and her child because a depressed mother may be less able to interact with her baby or respond in a warm and consistent manner. Unfortunately, her feelings about this can exacerbate and prolong her depression and she may express views that she is a bad mother. It is easy to see how such self-criticism makes it difficult for the new mother and those around her to cope with the depression.
Given that interventions for the mother will also help the well-being of the child there are many clinical programmes aimed at the early recognition and treatment of post-natal depression. Many obstetric and midwifery services use screening questionnaires to try to detect the problem as early as possible. This type of work has identified some very important issues with regard to the timing of onset of depressive symptoms. Contrary to the assumption that all women who wanted to become pregnant will be very content and happy throughout, it seems that many of the symptoms of so-called post-natal depressions can actually begin during the antenatal period. This finding has significant implications for the support and care offered to pregnant women and suggests that screening programmes need to commence earlier.
As discussed in Chapter 4, if a pregnant woman begins to experience depressive symptoms in the antenatal period it is likely that her stress hormone system is more active. Furthermore, the direct connections between the mother and baby (via the placenta) means that this hormonal overactivity can in some cases have an effect on the child’s responses to stress in their early life (because the hormones can cross the placenta and influence the development of the stress hormone system of the infant). Taken together this research emphasizes that treating depression associated with pregnancy and childbirth is important for both short-term and long-term well-being of the woman and the child.
Puerperal or post-partum psychosis is an uncommon condition, occurring in about one in 1,000 pregnancies. It is thought to be related to bipolar disorder and the symptoms may be accompanied by loss of reality or psychotic symptoms (hallucinations and delusions), and frequently by ideas of suicide or fear of harming the baby. The problem has been well recognized throughout history and the first psychiatric descriptions are attributed to Osiander in 1797. A description of the condition by Gooch that was written in the 1830s gives a flavour of the disorder: ‘the patient swears, bellows, recites poetry, talks bawdy and kicks up such a row that there is the devil to pay in the house’.
In a paper on the history of psychiatry, Hilary Marland provides an elegant review of the case notes of women diagnosed as suffering from puerperal insanity in asylums in the 19th century. Marland reports that many different stages of puerperal mania were described including states which are no longer recognized, such as dullness and relapse into drollery. Importantly, it is clear that the illness was often viewed in a judgemental way, being regarded as associated with sexuality and contravention of codes of decent female behaviour and maternal duty. Puerperal Insanity was attributed to the physical nature of childbirth but also to social factors such as poverty, poor nutrition, difficult family relationships, and stress. Treatment included feeding to the point of stoutness as well as rest and nutrition. In English literature, one of the most famous short stories thought to describe puerperal psychosis is The Yellow Wallpaper by Charlotte Perkins Gilman. The text has frequently been debated both for its depiction of the illness experience, but also for the conscious or subconscious maltreatment of the woman by her husband and his medical colleagues.
Today, puerperal psychosis is recognized as an extremely serious illness often regarded as a medical emergency requiring inpatient treatment at specialist mother and baby units. Reasons for the high level of concern about this problem become apparent through reading several of the published confidential inquiries into maternal deaths (defined as deaths that occur during pregnancy or in the year after the birth of the baby). Documents such as Why Mothers Die highlight that tragically suicide is the leading cause of death in new mothers. It is also the commonest cause of mothers killing their children usually in the heart-breaking belief that they are saving their child from future suffering (see Box 4).
Box 4 Infanticide
In the 19th century a plea of puerperal insanity was used as a defence in cases of infanticide and it was seen as a major public health problem in Europe at the time. To this day, peri-natal depression or puerperal psychoses remain the most common diagnoses associated with infanticide and if recognized by the court are usually associated with a more lenient penalty than other forms of murder.
Depression in men
Depression has so often been presented as a ‘woman’s disease’ that it is only relatively recently that health promotion and public information campaigns have recognized the need to target messages at the male population in order to improve the identification of depression and to increase the uptake of treatment in men. There are few differences in the nature of the symptoms experienced by men and women who are depressed, but there may be gender differences in how their distress is expressed or how they react to the symptoms. For example, men may be more likely to become withdrawn rather than to seek support from or confide in other people, they may become more outwardly hostile and have a greater tendency to use alcohol to try to cope with their symptoms. It is also clear that it may be more difficult for men to accept that they have a mental health problem and they are more likely to deny it, delay seeking help, or even to refuse help.
There are no reasons for the onset of depression that are unique to men, but some life events do seem to be particularly associated with the development of the problem. For example, becoming unemployed, retirement, and loss of a partner and change of social roles can all be risk factors for depression in men. In addition, chronic physical health problems or increasing disability may also act as a precipitant.
The relationship between physical illness and depression is complex. When people are depressed they may subjectively report that their general health is worse than that of other people; likewise, people who are ill or in pain may react by becoming depressed. Certain medical problems such as an under-functioning thyroid gland (hypothyroidism) may produce symptoms that are virtually indistinguishable from depression. Overall, the rate of depression in individuals with a chronic physical disease is almost three times higher than those without such problems. Evidence shows that depression is associated with an increased risk of developing certain conditions more than others, for example coronary heart disease, stroke, some cancers, and certain types of diabetes. These findings have become increasingly important as research indicates that some of these problems have shared genetic risk factors. Clinically, physicians and psychiatrists now recognize that the outcome of these physical disorders may be improved by treating the depression as well as the medical condition, and many programmes for chronic physical disorders now take this issue into account.
Suicide
A detailed account of the complexity of the underlying causes of suicide, clinical assessment of the risk of suicide, and its management is unrealistic in this short publication. However, it is impossible to write about mood disorders without acknowledging that individuals who are depressed are more likely than any other group in society to kill themselves. In this brief discussion we highlight some of the difficulties in assessing rates of suicide, what is known about current rates of suicide, offer a few comments on controversies (such as suicide rates and economic recession, copycat suicides, etc.), and what strategies actually reduce overall rates of suicide in a population.
A long-standing problem in gathering data about suicide is that many religions and cultures regard it as a sin or an illegal act. This has had several consequences. For example, coroners and other public officials often strive to avoid identifying suspicious deaths as a suicide, meaning that the actual rates of suicide may be under-reported. Also, in countries where suicide is illegal, criminalizing the act means that those left behind often experience further distress and stigmatization. Attitudes have begun to change, but the taboo about this subject has created problems about the collection of data and understanding of the reasons for suicide. Furthermore, perceptions of what constitutes suicide are also being reviewed and debates about assisted dying for those with terminal illnesses and the ‘right to die’ serve to demonstrate that this is an emotive issue that is always likely to provoke both sympathy and controversy.
According to the World Health Organization one person commits suicide every minute throughout the world; which equates to about one million people annually. All mental disorders carry an increased risk of premature death, but the risks are highest in depression and bipolar disorder where suicide is increased fifteen- to twentyfold compared to the general population. Rates of suicide have increased significantly in the last half-century, but it is also clear that they vary greatly between countries. The lowest annual rates are reported in Muslim and Latin American countries (about 6 suicides per 100,000 persons) and the highest in countries that were previously identified as Eastern Europe (about 30 per 100,000). Men die more often from suicide than women and men also tend to use more violent methods of suicide such as hanging or shooting, whereas women are more likely to take an overdose of medication.
Risk of suicide varies across the lifespan and the two age groups at highest risk are 15–24-year-olds and over 65-year-olds. In Western countries there has been a substantial increase in suicide rates in young men and it is thought that this may be secondary to access to lethal methods such as car exhaust fumes, high alcohol consumption, and lack of access to support or timely help and unemployment. There is some research that shows fluctuations in suicide rates according to variations in economic prosperity and recession, and a recent publication suggested that there were 10,000 additional suicides (over what would be predicted) associated with the recent economic recession in Europe. These findings have some parallels with early theories of suicide that highlight the influence that social factors may have on individuals.
In 1897, Émile Durkheim, a French sociologist, published his study of suicide that argued that the causes of suicide were linked more with social factors than individual characteristics. He observed that the rate of suicide varied with time and place, for example being less common during times of peace than during times of war and more likely in times of economic depression rather than prosperity. He looked for factors that explained variations other than emotional stress, such as the degree to which individuals feel integrated into society and developed a typology to describe different forms of suicide (see Box 5).
Box 5 Durkheim’s typology of suicide
Most individuals with depression can and do recover from their illness episode and, even if they experience further relapses, suicide is a rare outcome. Nevertheless, suicide is viewed as a tragic consequence that should be prevented if at all possible and so clinical programmes have been introduced to improve the detection of depression and provide early access to effective treatments especially for vulnerable groups, such as people recently discharged from a psychiatric hospital. These strategies are combined with training to ensure that clinicians ask depressed patients if they have thought about harming themselves. Suggestions that asking such questions would increase the likelihood that the patient would act on ideas of suicide are entirely unfounded, and indeed for most patients being able to talk to a professional about these thoughts is a relief. Whilst ensuring that clinicians identify individuals at increased risk of suicide is vitally important, research suggests that ultimately, the most effective way of decreasing the rates of suicide is to use population-based interventions ranging from guidelines on media reporting through to reducing access to means.
It is difficult to be certain whether the media portrayal of suicide might lead to someone else killing themselves. Evidence for copycat suicides is equivocal, although it is suggested that the risk is especially increased when a famous person is involved. However, concerns about the risk of copycat suicides are not new, and there are documented examples from previous centuries. For example, there was a spate of copycat suicides in 1774 after the publication of Goethe’s novel Die Leiden des Jungen Werthers (The Sorrows of Young Werther), which described how a young man killed himself because he was unlucky in love. The book was eventually banned.
More recently, concerns have been voiced that some suicides may have occurred after individuals became distressed as a result of cyber-bullying which includes negative or abusive comments being posted on internet web-pages or circulated via different internet sites. This has led to some attempts to promote guidelines on responsible reporting of suicides in the media and attempts to regulate access to websites or modify the content of unregulated sites (although the latter is difficult to implement).
The most effective strategy for reducing suicide rates is to reduce access to methods for killing oneself. For example, suicide rates are lower in countries with stricter limitations on access to firearms. Also, in the early 20th century placing one’s head in a gas oven was a common method of suicide in the United Kingdom, but following the conversion from coal gas to North Sea gas the suicide rate dropped. Following on from this, new car exhausts are now fitted with catalytic converters to reduce deaths by carbon monoxide poisoning.
Restricting access to barbiturates in the 1960s led to a 23 per cent reduction in suicide by these drugs; limiting the purchase of over-the-counter painkillers (analgesics) such as paracetamol and using blister packs to slow ingestion have been found to be effective in some studies. Other public health interventions include the erection of barriers or nets placed in suicide hotspots such as the Clifton Suspension Bridge in Bristol in the United Kingdom and the Golden Gate Bridge in San Francisco in the USA (see Figure 3). Telephone helplines such as the Samaritans provide the opportunity for people to talk in confidence in the hope this will prevent them from acting on suicidal thoughts and signs displaying telephone helpline numbers are displayed on many high-level bridges.
3. The crisis counselling sign on Golden Gate Bridge in San Francisco.