Cookies on this website

We use cookies to ensure that we give you the best experience on our website. If you click 'Accept all cookies' we'll assume that you are happy to receive all cookies and you won't see this message again. If you click 'Reject all non-essential cookies' only necessary cookies providing core functionality such as security, network management, and accessibility will be enabled. Click 'Find out more' for information on how to change your cookie settings.


The Oxford Monitoring System for Attempted Suicide was established in 1976. It has resulted in many publications on self-harm and provides regular reports for local clinicians. Studies published using local data collected as part of the multicentre study include:


Haw, C., Casey, D., Holmes, J. & Hawton K. (2015). Suicidal intent and method of self-harm: a large-scale study of self-Harm patients presenting to a general hospital. Suicide and Life Threatening Behaviour Apr 27. Link to the paper.

We used information from the Oxford Monitoring System for Self-harm for 2004 to 2011 to study 4,840 hospital presentations for self-harm in which patients were scored on the Suicidal Intent Scale (a measure of the extent to which an individual seemed to want to die). Higher suicide intent scores found in males and were associated with increasing age, self-poisoning versus self-injury, multiple methods of self-harm versus self-injury alone, use of gas (mainly carbon monoxide), dangerous methods of self-injury (including hanging, gunshot), and use of alcohol as part of the act. In patients who took overdoses, higher suicide intent was found where more tablets were taken. There was relatively little difference in suicide intent scores between overdoses of different drugs. Use of alcohol within 6 hours of self-harm was associated with lower suicide intent score scores. 
Conclusions: Certain methods of self-harm, particularly dangerous methods of self-injury and self-poisoning with gas, were associated with high intent and should alert clinicians to potential higher risk of suicide. However, apart from use of gas, it is important for clinicians to note that a patient’s level of suicidal intent cannot be inferred from the type of drug used for self-poisoning. 

Hawton, K., Haw, C., Casey, D., Bale, L., Brand, F. & Rutherford, D. (2015). Self-harm in Oxford, England: epidemiological and clinical trends, 1996-2010. Social Psychiatry and Psychiatric Epidemiology 50(5), 695-704. Link to the paper.

We analysed data on all self-harm presentations to the general hospital in Oxford between 1996 and 2010 using the Oxford Monitoring System for Self-harm in order to investigate trends in prevalence, methods and repetition of self-harm, and receipt of psychosocial assessment by a mental health specialist. Psychosocial assessment includes investigation of patient’s problems and needs, together with future risks. For patients receiving a psychosocial assessment, we investigated trends in alcohol use and misuse, prior psychiatric treatment, problems, and suicidal intent. Rates of self-harm rose in both genders between 1996 and 2002/2003, after which they declined. There was evidence of a possible cohort effect, whereby higher rates in younger males in earlier years transferred over time to older age groups. Self-cutting, hanging and jumping became more common. Paracetamol ingestion was a common method of self-poisoning. Overdoses of antidepressants also increased. Alcohol use in relation to self-harm, and also alcohol-related problems, became more common. History of prior psychiatric treatment and, especially, of prior self-harm, also increased from 2008, as did the proportion of patients with employment problems. Despite national guidance on the need for psychosocial assessment the proportion of patients receiving an assessment declined.
Conclusions: Major changes in the extent and nature of self-harm occurred over the study period, some suggestive of increased psychiatric problems, and others reflecting prescribing practices and changes in drinking patterns. The findings emphasise the need for psychosocial assessment following self-harm to identify treatment needs and reduce repetition. 


Hiles, S., Bergen, H., Hawton, K., Lewin, T., Whyte, I. & Carter, G. (2015). General hospital-treated self-poisoning in England and Australia: Comparison of presentation rates, clinical characteristics and aftercare based on sentinel unit data. Journal of Psychosomatic Research 78(4), 356-362. Link to the paper.

Hospital-treated deliberate self-poisoning is common but existing national monitoring systems are often deficient. Clinical practice guidelines in the UK and Australia recommend universal psychosocial assessment of patients presenting to general hospitals. In this study we compared presentation rates, patient characteristics, psychosocial assessment and aftercare in self-poisoning patients in England and Australia. We used data collected in Oxford in England (3,042 patients) and Newcastle in Australia (3,492 patients) between 1997 and 2006. The rates of presentation for self-harm were considerably higher in Oxford than Newcastle in both males and females. In both centres approximately 70% of presentations occurred outside the normal working day. Most patients (95%) were admitted to a bed in a general hospital. There were differences in the pattern of drugs used for self-poisoning, although paracetamol, minor tranquilisers and antidepressants were the most frequent. Co-ingestion of alcohol occurred in 24% of patients in Oxford and 32% in Newcastle. Ninety three per cent of patients in Newcastle received a psychosocial assessment in hospital compared with 80% in Oxford. A far higher proportion of patients were transferred to psychiatric inpatient care after discharge from the hospital in Newcastle than in Oxford.
Conclusions: This is the first study we are aware of comparing self-poisoning between centres in the UK and in Australia. Rates of self-poisoning appear to be higher in the UK than in Australia, although many other characteristics of patients are quite similar. A vast majority of self-poisoning patients received a psychosocial assessment while in hospital but there were differences in aftercare, especially inpatient psychiatric admission. This study shows that sentinel monitoring of routine care of self-poisoning patients can provide valuable comparisons between countries. 

Meyer, N., Voysey, M., Holmes, J., Casey, D. & Hawton K. (2014). Self-harm in people with epilepsy: a retrospective cohort study. Epilepsia 55(9), 1355-65. Link to the paper.

Little is known about self-harm in people with epilepsy, despite suicide being recognized as an important cause of death in this population. We investigate the characteristics of self-harm in people with epilepsy who presented to the general hospital in Oxford following self-harm between 1994 and 2008. Epilepsy diagnosis was confirmed through review of medical records. We compared the characteristics of 132 people with epilepsy and 9,778 self-harm patients without epilepsy (after accounting for age, sex, and repetition of self-harm). We also identified deaths that occurred up until the end of 2011 in patients presenting between 1998 and 2008. The number of episodes of self-harm per individual with epilepsy was approximately double that of other self-harm patients and the time between first and second self-harm events was shorter. People with epilepsy were significantly more likely to use antiepileptic medication in overdose, although overall methods of self-harm were similar in the two groups. We found no major differences in suicide intent scores (a measure of the extent to which an individual seemed to want to die) or the proportion of patients who subsequently died by suicide. Previous outpatient psychiatric treatment, longer duration of unemployment, experience of violence, and housing problems were associated with self-harm in people with epilepsy.
Conclusions: People with epilepsy who self-harm do so more frequently than other self-harm patients. Clinicians should be aware of this and pay attention to contributory factors as these may enhance risk in this population. 


Haw, C., Hawton, K. (2011) Problem drug use, drug misuse and deliberate self-harm: trends and patient characteristics, with a focus on young people, Oxford, 1993-2006. Social Psychiatry and Psychiatric Epidemiology, 46, 85-93. Link to the paper.

Drug misuse is related to self-harm and suicide. However, relatively little is known about self-harm in patients with drug problems and whether drug misuse by self-harm patients is increasing. We used data collected by the Oxford Monitoring System for Attempted Suicide to study the characteristics of self-harm patients with drug problems who presented to the general hospital in Oxford between 1993 and 2006, and who underwent psychosocial assessment at their first presentation in the study period. We also studied trends in problem drug use and drugs misused over this period. During the 14-year study period, 11,426 patients presented of whom 9,248 underwent psychosocial assessment and it was known whether or not they had a drug problem. Problem drug use was present in 805/9,248 (8.7%) patients. Problem drug use was more common in males (13.6%) than in females (5.3%). Problem drug users were younger, more likely to be socially disadvantaged, to have a personality disorder and co-morbid alcohol problems, and to have a further episode of self-harm within a year. Problem drug use in young females was associated with higher suicidal intent scale (SIS) scores. During the study period, problem drug use and drug misuse increased in females, but not in males. Cannabis and cocaine misuse increased with time. 

Conclusions: Provision of help for self-harm patients with problem drug use is particularly challenging due to their complex social and clinical characteristics and increased risk of further self-harm, suicide and accidental death. It may require extensive liaison between different services. The increasing misuse of drugs by females who self-harm and the higher suicidal intent of young females are of concern. 

Harriss, L. & Hawton, K. (2011). Deliberate self-harm in rural and urban regions: A comparative study of prevalence and patient characteristics. Social Science & Medicine. Link to the paper.

In countries like the UK, people living in urban regions are more likely to suffer poor physical and mental health than rural populations, and to have increased rates of psychiatric disorder. Urban/rural differences in suicidal behaviour have most frequently focussed on variations in the occurrence of suicide. We investigated rates of self-harm in urban and rural districts of Oxfordshire, England, and compared characteristics of self-harm patients resident in these two areas. Information was collected on 6833 self-harm episodes by 4054 persons aged 15 years and over presenting to the local general hospital between 2001 and 2005. We found that rates of self-harm in urban areas were substantially higher than those in rural areas amongst both males and females aged between 15 and 64 years. This difference was sustained even when measures of socio-economic deprivation and social fragmentation were taken into account. There was little difference between urban and rural rates of self-harm for patients aged 65 years and over. People who self-harmed in urban areas were more likely to be younger, non-white in ethnic origin, unemployed, living alone, to have a criminal record, to have previously engaged in DSH, and to report problems with housing. Self-harm patients from rural areas were more likely to have suffered from physical illnesses, and to have higher suicide intent scores.
Conclusions: Higher rates of self-harm were found for urban areas in people aged 15-64 years. There were also differences in characteristics of self-harm patients from urban compared with rural areas. Results of studies such as this can help identify where resources for preventive initiatives should be primarily directed and also what types of individuals may be at most risk in different areas. However, since variation by area will in part be due to differences at the individual level, further research utilising more sophisticated multi-level modelling techniques would be useful to determine the relative contribution of area and individual characteristics to relative risks of self-harm in urban and rural areas. 


Mahadevan,S., Hawton, K., Casey, D. (2010) Deliberate self-harm in Oxford University students, 1993-2005:A descriptive and case-control study. Social Psychiatry and Psychiatric Epidemiology, 45, 211-219. Link to the paper.

Self-harm is a major problem among young people and has been identified as one of the key mental health problems affecting students. We analyses data on self-harm presentations to the general hospital in Oxford by Oxford University students for the 12-year period 1993 to 2005. The characteristics of the students who self-harmed were compared with those of age-matched individuals from Oxford City. Problems with academic work, relationships with family, partners and friends were most likely to contribute to self-harm episodes in students. Many experienced problems with psychiatric disorders and social isolation. The frequency of eating disorders was very high in students, and contributed to self-harm significantly more often than in the comparison population. Fewer students self-poisoned in the self-harm episode, fewer had personality disorder and fewer had problems with physical health, finance, housing and violence. Alcohol consumption in association with self-harm and alcohol-related problems were common in both students and the comparison population. Male students had significantly higher suicide intent than the other males who self-harmed. Many students were referred to the university counselling service for follow-up (a resource not available to non-students).
Conclusions: Comparison of university students who self-harmed with other individuals of the same age who self-harmed showed key differences in psychiatric characteristics, problems contributing to DSH and aftercare offered. These findings may help in the design of targeted self-harm prevention and management strategies for students. 

Hawton, K., Bergen, H., Mahadevan, S., Casey, D., Simkin, S. (2010) Suicide and deliberate self-harm in Oxford University students over a 30-year period. Social Psychiatry and Psychiatric Epidemiology, Link to the paper. 

In the past, rates of suicide in Oxford University students have been reported to be high. In this study, the largest of its kind, we aimed to determine whether rates of suicide and self-harm in university students differ from those in other young people. We obtained information on Oxford University students who died by suicide or presented to hospital following self-harm between 1976 and 2006 from official records and a general hospital monitoring system in Oxford. Rates of suicide and self-harm in the students and in other young people in the general population were calculated from university, local and national population figures. Forty-eight Oxford University students (32 males, 16 females) died by suicide over the 30-year period. Most (N=42) were aged 18-25 years. The suicide rate did not differ from that of other people in this age group in England and Wales. There was evidence of clustering of specific methods of suicide over time. During the same period 602 students (383 females and 219 males) presented to the general hospital following self-harm. Most (90.7%) were aged 15-24 years, in which age group rates of self-harm during term-time were lower than in other young people in Oxford City. There was an excess of student self-harm episodes in the main exam term.
Conclusions: Contrary to earlier findings and popular belief, suicide rates in Oxford University students do not differ from those in other young people. Rates of self-harm are much lower than in other young people. Risk of self-harm may increase around the time of examinations. 





The Manchester Self-Harm (MaSH) Project is a clinical audit and research project established at the University of Manchester and forms part of the Centre for Mental Health and Risk. The purpose of the MaSH Project is to inform local services about self-harm presentations to the Emergency Department (ED) and to contribute to national research. Studies published using local data collected as part of the multicentre study.

Clements, C., Jones, S., Morriss, R., Peters, S., Cooper, J., While, D. & Kapur, N (2014). Self-harm in bipolar disorder: Findings from a prospective clinical database. Journal of Affective Disorders 173, 113-119. Link to the paper.

People with bipolar disorder are thought to be at high risk of engaging in suicidal behaviours, including self-harm. How that risk presents, and which people with bipolar disorder are most at risk remains in question. We looked at data on all people with a primary diagnosis of bipolar disorder who presented to hospitals in Manchester following self-harm (n=103). These people were compared to people without bipolar disorder who had also presented to hospital following self-harm (n=515). People with bipolar disorder repeated more often, and sooner, than other people who self-harmed. They were also more likely to have a history of previous self-harm and to be in contact with psychiatric services at the time of the self-harm presentation.
Conclusions: In this study we found that people with bipolar disorder who self-harm have a higher risk of repetition than people who self-harm more generally. These results suggest that good quality monitoring and responsive management of symptoms by psychiatric services could help to reduce self-harm in people with bipolar disorder. 


Chang, S., Steeg, S., Kapur, N., Webb, R., Yip, P. & Cooper, J (2015). Self-harm amongst people of Chinese origin versus White people living in England: a cohort study. BMC Psychiatry 15(1), 79. Link to the paper.

There has been little previous research on self-harm among people of Chinese origin living in the UK, although this population has grown substantially in recent years and China is now the largest source of international students at UK universities. Using a study cohort of people presenting to hospital following self-harm in Manchester, we compared self-harm rates in people of Chinese ethnic origin (n=45) with rates in White people (n=7,111). Our study showed that people of Chinese origin in the UK had less than one fifth the incidence of self-harm compared with White people, and incidence was particularly low amongst men in this ethnic group. Compared with White people, people of Chinese origin who self-harmed were younger and more often female and students. They were more likely to describe relationship problems as a precipitant for self-harm. They were also more likely to self-injure but less likely to self-poison with drugs.
Conclusions: In this study we found lower rates of self-harm in people of Chinese ethnic origin than in White people. Future research is required to understand this low incidence, and to investigate whether this masks hidden episodes within the community, potentially due to less help-seeking or more barriers to accessing services. Healthcare professionals need to be aware of the risk characteristics of people of Chinese origin who self-harm.


Murphy, E., Kapur, N., Webb, R. & Cooper, J (2011). Risk assessment following self-harm: comparison of mental health nurses and psychiatrists. Journal of Advanced Nursing, 67, 127-39. Link to the paper.

Specialist psychosocial assessments following self-harm are carried out by either mental health nurses or psychiatrists. An essential component of this is an assessment of the risk of self-harm repetition. We compared nurses’ and psychiatrists’ risk assessments and found them highly similar in terms of their ability to predict which patients would subsequently repeat self-harm. They were also in strong agreement regarding the risk factors they used to inform their assessments. However, they differed markedly in terms of the aftercare they recommended. Psychiatrists were over 5-times as likely to request psychiatric inpatient admission, whereas nurses more frequently made outpatient referrals.
Conclusions: The results of this study on risk assessment support the provision of nurse-led assessment services. However, the referral differences between doctors and nurses have important implications for equity of patient care. Multidisciplinary approaches to risk management training may help to reduce variations in referral practices. 


Oude Voshaar, R., Cooper, J., Murphy, E., Steeg, S., Kapur, N. & Purandare, N (2011). First episode of self-harm in older age: a report from the 10-year prospective Manchester Self-Harm project. J Clin Psychiatry, 72, 737-43. Link to the paper.

Older people who self-harm are more likely to die by suicide than the younger age groups. There is very little research on older people who self-harm, particularly if it is their first attempt. When we compared the risk of repetition of first-ever self-harm in later life (55 years and over) to first-ever self-harm in middle age patients (35-54 years) we found older people were less likely to repeat although repetition was more often fatal amongst the older group. Their circumstances suggest higher suicidal intent at the time of self-harm in older people. An important factor in repetition in older people was physical health problems, whereas psychiatric characteristics had little impact on the risk of repetition in old age.
Conclusions: High suicidal intent and different predictors of repetition in first-ever self-harm in older age highlight the need for age-specific interventions beyond the scope of psychiatric care alone, and include rehabilitation programmes for those who present with self-harm precipitated by a physical disease. 

Murphy, E., Steeg, S., Cooper, J., Chang, R., Turpin, C., Guthrie, E. & Kapur, N (2010). Assessment rates and compliance with assertive follow-up after self-harm: cohort study. Arch Suicide Res, 14, 120-34. Link to the paper.

After presenting to hospital with self-harm, some patients are either not offered an assessment, or do not attend offered assessment or treatment. We studied a specialist self-harm team who assertively follow-up these patients in their homes. Rates of assessment, offers and completion of therapy ranged from 50% to 60%. These results may reflect some of the challenges of engaging this patient group. However, of those offered therapy, 73% attended at least one session. Attendance was higher for clients who had more depressive symptoms and who were also receiving treatment with their GP. 
Conclusions: People who leave hospital without assessment may be a difficult to engage group. Primarily, emergency departments should aim to assess patients while they are still in hospital. Subsequently, specialist community self-harm teams may provide a means for following-up those who leave without assessment. 


Dickson, S., Steeg, S., Gordon, M., Donaldson, I., Matthews, V., Kapur, N., & Cooper, J (2011). Self-Harm in Manchester 2008-2009. The Centre for Suicide Prevention: The University of Manchester. Link to the report.

Dickson, S., Steeg, S., Donaldson, I., Matthews, V., Healey, M., Cooper, J., Kapur, N., & Murphy, E. (2009). 'Self-Harm in Manchester' (01:09:05-31:08:07). Manchester: The University of Manchester.

Studies by the MaSH Project prior to 2008 are listed in full at  the following link.

Derby Mental Health Liaison Team (MHLT) (2011). Derby Annual Report of 2010 attendances.