Responding to the rising prevalence of self-harm

Young people who self-harm have substantially increased risk of adverse non-fatal and fatal outcomes, including suicide, compared with those who do not self-harm. The frequency of young people presenting to general practices and emergency departments after self-harm has increased in the past 20 years in England and Australia. However, people who present to medical services after self-harm differ from those who do not, and most people who self-harm do not seek medical attention. Thus, whether this increase in presentation reflects a true rise in the prevalence of self-harm remains unclear. In The Lancet Psychiatry, Sally McManus and colleagues address this uncertainty. The authors analysed data for non-suicidal self-harm (NSSH) from three waves of the Adult Psychiatric Morbidity Survey administered to people aged 16 years or older in the general population of England, Scotland, and Wales in 2000, 2007, and 2014. To maximise comparability between waves (which differed slightly in terms of locations and the age range of participants included), they focused on data for people aged 16–74 years who lived in England. Notable strengths of the study include the repeat probability-sample survey design, the population sampling frame, and the inclusion of respondents not in contact with mental health services (who are often excluded from studies of NSSH). McManus and colleagues’ report that the prevalence of lifetime NSSH increased from 2·4% (95% CI 2·0–2·8) to 6·4% (5·8–7·2) between 2000 and 2014, with rises noted in both male and female participants and across all age groups. The largest increase was in women and girls aged 16–24 years (from 6·5% [95% CI 4·2–10·0] to 19·7% [15·7–24·5] in 2014). Data from the 2014 survey suggest a concurrent rise in anxiety and depression in girls and young women, which could partly explain the increase in NSSH. With one in five female 16–24-yearolds in England reporting lifetime NSSH in 2014—a proportion higher than the pooled global prevalence estimate of past-year common mental disorder (17·6%)—NSSH in young people is clearly an important public health issue. Despite the increased prevalence of NSSH, McManus and colleagues noted no significant change in the proportion of respondents who subsequently sought support from health services. At each wave, more than 50% of respondents who reported lifetime NSSH sought no medical or psychological treatment and, in 2014, such help-seeking was even less likely in male respondents and people younger than 35 years. Coupled with the rise in NSSH, these findings underscore the importance of both population-level and targeted prevention strategies to reduce the incidence of NSSH, complemented by efforts to promote help-seeking from low-threshold mental health services for people who engage in NSSH. NSSH is a complex, multifaceted behaviour with myriad precipitants and functions. McManus and colleagues also provided informative data about the motivations driving NSSH. The prevalence of NSSH to cope with unpleasant emotions (including anger, tension, anxiety, and depression) roughly tripled in both sexes between 2000 and 2014. This was the most common reason provided for engaging in NSSH, and was reported by 17·7% ([95% CI 13·9–22·3]) of women and girls aged 16–24 years in the 2014 survey. Because most NSSH does not lead to help-seeking, populationlevel interventions to enhance emotion regulation and resilience in young people, including through schoolbased initiatives, could help to reduce the incidence of NSSH. NSSH is difficult to assess accurately. In addition to the limitations associated with assessing NSSH via self-report, the lifetime measure of NSSH used by McManus and colleagues did not take into account the frequency or recency of NSSH, meaning that the temporal variability of NSSH during adolescence was not captured. Additionally, despite the purported population sampling frame, people living in institutional settings, such as prisons, youth detention centres, and psychiatric hospitals, were excluded. Thus, the prevalences reported by McManus and colleagues are probably underestimates, because NSSH is common in institutional settings. Young people in such settings are typically excluded from whole population studies, despite the high mental health burden. Inclusion of these highly vulnerable young people in research and policy responses needs to form part of a comprehensive strategy to prevent NSSH at the population level. Finally, the As hl ey C oo pe r/ Sc ie nc e


Responding to the rising prevalence of self-harm
Young people who self-harm have substantially increased risk of adverse non-fatal and fatal outcomes, including suicide, compared with those who do not self-harm. 1, 2 The frequency of young people presenting to general practices and emergency departments after self-harm has increased in the past 20 years in England 3 and Australia. 4 However, people who present to medical services after self-harm differ from those who do not, 5 and most people who self-harm do not seek medical attention. 5 Thus, whether this increase in presentation reflects a true rise in the prevalence of self-harm remains unclear.
In The Lancet Psychiatry, Sally McManus and colleagues 6 address this uncertainty. The authors analysed data for non-suicidal self-harm (NSSH) from three waves of the Adult Psychiatric Morbidity Survey administered to people aged 16 years or older in the general population of England, Scotland, and Wales in 2000, 2007, and 2014. To maximise comparability between waves (which differed slightly in terms of locations and the age range of participants included), they focused on data for people aged 16-74 years who lived in England. Notable strengths of the study include the repeat probability-sample survey design, the population sampling frame, and the inclusion of respondents not in contact with mental health services (who are often excluded from studies of NSSH). McManus and colleagues' report that the prevalence of lifetime NSSH increased from 2·4% (95% CI 2·0-2·8) to 6·4% (5·8-7·2) between 2000 and 2014, with rises noted in both male and female participants and across all age groups. The largest increase was in women and girls aged 16-24 years (from 6·5% [95% CI 4·2-10·0] to 19·7% [15·7-24·5] in 2014). Data from the 2014 survey 7 suggest a concurrent rise in anxiety and depression in girls and young women, which could partly explain the increase in NSSH. With one in five female 16-24-yearolds in England reporting lifetime NSSH in 2014-a proportion higher than the pooled global prevalence estimate of past-year common mental disorder (17·6%) 8 -NSSH in young people is clearly an important public health issue.
Despite the increased prevalence of NSSH, McManus and colleagues noted no significant change in the proportion of respondents who subsequently sought support from health services. At each wave, more than 50% of respondents who reported lifetime NSSH sought no medical or psychological treatment and, in 2014, such help-seeking was even less likely in male respondents and people younger than 35 years. Coupled with the rise in NSSH, these findings underscore the importance of both population-level and targeted prevention strategies to reduce the incidence of NSSH, complemented by efforts to promote help-seeking from low-threshold mental health services for people who engage in NSSH.
NSSH is a complex, multifaceted behaviour with myriad precipitants and functions. McManus and colleagues also provided informative data about the motivations driving NSSH. The prevalence of NSSH to cope with unpleasant emotions (including anger, tension, anxiety, and depression) roughly tripled in both sexes between 2000 and 2014. This was the most common reason provided for engaging in NSSH, and was reported by 17·7% ([95% CI 13·9-22·3]) of women and girls aged 16-24 years in the 2014 survey. Because most NSSH does not lead to help-seeking, 5 populationlevel interventions to enhance emotion regulation and resilience in young people, including through schoolbased initiatives, could help to reduce the incidence of NSSH.
NSSH is difficult to assess accurately. In addition to the limitations associated with assessing NSSH via self-report, the lifetime measure of NSSH used by McManus and colleagues did not take into account the frequency or recency of NSSH, meaning that the temporal variability of NSSH during adolescence 9 was not captured. Additionally, despite the purported population sampling frame, people living in institutional settings, such as prisons, youth detention centres, and psychiatric hospitals, were excluded. Thus, the prevalences reported by McManus and colleagues are probably underestimates, because NSSH is common in institutional settings. 10 Young people in such settings are typically excluded from whole population studies, despite the high mental health burden. Inclusion of these highly vulnerable young people in research and policy responses needs to form part of a comprehensive strategy to prevent NSSH at the population level. Finally, the Comment www.thelancet.com/psychiatry Vol 6 July 2019 549 small numbers of people who self-harmed did not permit robust analysis by ethnic group. In view of the reported differences between ethnic groups in the prevalence and characteristics of NSSH, and in the frequency of subsequent service use, 11 complementary use of data-linkage methods-which are effective for studying inclusion health populations and stigmatised behaviours such as NSSH-might be beneficial in future studies. Adolescent NSSH is a marker for concurrent risk behaviours that pose substantial hazards for social and emotional development throughout young adulthood and beyond. 1 Interventions addressing other key individual-level risk domains-including mental health problems, substance use, and antisocial behaviourare therefore likely to help to support vulnerable young people as they make their way through life. 1 Many of the girls and young women in McManus and colleagues' study who reported NSSH will form part of the next generation of mothers in England. In a population-based intergenerational cohort study 12 in Australia, new mothers who self-harmed only in young adulthood (ie, when aged 20-29 years) scored twice as high on measures of postpartum depression and mother-infant bonding problems than those who self-harmed only during their teenage years and those with no history of NSSH. This finding suggests that, in addition to the established adverse health and social outcomes associated with NSSH during adolescence, 1 NSSH during young adulthood could also be a marker of future vulnerability to perinatal mental health and mother-infant bonding problems. The longer-term consequences of maternal NSSH for the health and wellbeing of offspring beyond the perinatal period are unknown.
Evidence for effective interventions to reduce NSSH in young people is scarce, and key studies require replication. 13 Thus, the increase in the prevalence of NSSH has wide-reaching and potentially lifelong implications. Sequelae of the growing burden of NSSH include a rise in preventable suicide deaths in vulnerable young people and a poorer start in life for the children of young women engaging in self-harm. Unless evidence-based, population-level, and targeted preventive responses are scaled up proportionately to the increase in NSSH, and access to low-threshold mental health services for young people engaging in NSSH is improved, the burden of untreated NSSH will continue to rise.