Timing of supports and interventions following self-harm is crucial.

Those who present to hospital following self-harm are one of the groups at greatest risk for suicide, with 1.6% dying by suicide within 12 months. In addition, a significant proportion will present with a further act of self-harm. In The Lancet Regional Health – Western Pacific, Yi Chai and colleagues present comprehensive data on predictors and patterns of repeat attendance to hospital with self-harm in Hong Kong, over a period of 15 years. Based on a cohort of 99,116 individuals – accounting for 127,801 hospital presentations – the reported 12-month risk of repetition after a self-harm presentation was 14.25% (95% CI: 14.04–14.46%). Male gender, older age, social welfare for payment, self-injury as a method, psychiatric admission and presence of depression, personality disorder or substance misuse all elevated risk of repetition. However the strongest predictor of subsequent repetition was self-harm history, with a five-fold risk for those with four or more previous attendances (HR: 4.81, 95% CI: 4.46–5.18). The study uses a rigorous methodology, linking multiple sources of health data and examining recurrent self-harm presentations over a long follow-up period. Record linkage allowed for a wide range of predictors to be examined, a limitation of many studies on this topic. Some findings may be specific to Hong Kong, including increasing rates of repetition over time. The negative association between self-harm repetition and alcohol misuse may also reflect general population trends.

Few countries have reliable data on self-harm and not many have focused on Asian populations when examining repetition. The reported 12-month risk of repetition is higher than previously reported for the region (approximately 10%) and more similar to estimates from European countries. The authors suggest that this may be due in part to the profile of methods of self-harm in Hong Kong, as well having less bias regarding loss to follow-up than other regional registries. Authors of a recent study from Sri Lanka suggest that low repetition rates may be a feature of low- and middle-income countries, reflecting a range of factors including a high case fatality rate of pesticide poisonings, longer inpatient stays following self-harm and a lower prevalence of mental disorders compared to Western and other high-income countries. As a consequence, country-specific data and prevalence estimates, which take into account contextual factors, are essential to inform health service responses and to monitor trends within regions and across time.

These findings reinforce the need to deliver appropriate interventions following self-harm, as well as standardised care at the time of attendance to hospital. There is some evidence for reduced rates of repetition following routine aspects of clinical management and brief interventions in the emergency department, as well as for psychosocial therapies. Still, there is a significant gap between what we know works and what is provided. Clinical management of self-harm in acute settings is inconsistent, likely reflecting policies and resources in each hospital. Similar to previous work, this study shows that those who present frequently are at greatest risk of repetition. Self-harm is an indication of significant ongoing distress for an individual. Yet despite this, those who attend frequently often report negative experiences and can be viewed negatively by emergency staff. A consistent approach to management of self-harm, regardless of self-harm history, including a comprehensive psychosocial assessment at each presentation, may result in a more positive experience for these patients. The findings from Chai and colleagues support existing knowledge of the risk of repetition being greatest in the days and weeks following a self-harm presentation, providing services with clear guidance concerning the provision of timely follow-up. This, and other evidence which suggests that suicide risk is greatest in the 12 months after a self-harm presentation, means that reducing the time for follow-up is a crucial target, requiring joined up efforts and co-ordination across the health service. There are also barriers to accessing evidence-based follow on care, which will vary by country. However initiatives which standardise and co-ordinate responses to self-harm are lacking, and few are evaluated in terms of implementation.

Large-scale epidemiological studies and registries of self-harm have a vital role in informing research as well as guiding policy on how best to improve how we respond to people who attend hospital in suicidal crisis and following self-harm. They identify potential gaps in clinical practice, including the availability and timing of interventions, the need for standardised care, as well as enabling monitoring of changes in profile and repetition over time. Studies which gather evidence on the development and implementation of integrated mental health services and programmes should be prioritised, in order to share learnings about best practice when supporting those who self-harm. Finally, the potential impacts of COVID-19 on mental health will likely compound existing challenges in accessing appropriate services. Alternative ways of providing support, for example through the delivery of digital interventions, need to be explored.

Declaration of Competing Interest

None to declare.