Cookies on this website
We use cookies to ensure that we give you the best experience on our website. If you click 'Continue' we'll assume that you are happy to receive all cookies and you won't see this message again. Click 'Find out more' for information on how to change your cookie settings.

Kapur, N., Steeg, S., Webb, R., Haigh, M., Bergen, H., Hawton, K. Ness, J. Waters, K. Cooper, J. (2013) Does Clinical Management Improve Outcomes following Self-Harm? Results from the Multicentre Study of Self-Harm in England. PLoS ONE 8(8): e70434. Link to the paper.

It is unclear from research to date what works and does not work in the management of self-harm and the treatments that are suggested in guidelines may not be available in routine practice. We used ten years of data from the Multicentre Study of Self-Harm in England to examine how the management that patients receive in hospital relates to their risk of repeat self-harm. We investigated the relationship between four aspects of management (psychosocial assessment, medical admission, psychiatric admission, referral for specialist mental health follow up) and repetition of self-harm within 12 months, adjusting results to account for differences in patients’ demographic and clinical characteristics. The study included 35,938 individuals presenting with self-harm between 2000 and 2009. Psychosocial assessment was associated with a 40% lower risk of repetition in two of the three centres, and this was the case even after taking into account other medical care and mental health follow-up arrangements. The protective effects of psychosocial assessment may have been weakest in people who lived in the most deprived areas.
These findings add to the evidence that thorough assessment is central to the management of self-harm. Further work is needed so that we can understand more about why assessment is linked to better outcomes for some patients but not for others and the possible mechanisms by which assessment is effective.

 

Steeg, S., Kapur, N., Webb,R., Applegate, E., Stewart, S.L.K., Hawton,K., Bergen, H., Waters, K., Cooper, J. (2012) The development of a population-level clinical screening tool for self-harm repetition and suicide: the ReACT Self-Harm Rule. Psychological Medicine, 7:1-12. Link to the paper.

Clinical decision tools have been used to inform clinical management in various areas of medicine. In this study we aimed to develop a clinical tool to help identify patients at higher risk following a self-harm presentation to the Emergency Department. We developed the tool using data from Manchester and Derby and tested it separately on data from Oxford. We found four factors that classified the self-harm presentation as higher risk of repeat self-harm, or suicide, within 6 months: recent self-harm (in the past year), living alone or homelessness, cutting as a method of harm and treatment for a current psychiatric disorder. These factors comprise the ReACT Self-Harm Rule. In Manchester and Derby, the tool performed with 95% (CI 94 to 95%) sensitivity (the proportion of repeat SH attendances or suicides correctly identified as higher risk) and 21% (21 to 22%) specificity (the proportion of non-repetitions that were correctly classified as low risk). There were 92 subsequent suicides in the three centres, of which 83 were classed as higher risk by the tool. 
The ReACT Self-Harm Rule might be used as a screening tool to inform the process of assessing SH presentations to ED. The four risk factors could also be used as an adjunct to in-depth psychosocial assessment to help guide risk formulation. The tool should not replace detailed psychosocial assessment of risk and needs. Decisions on follow-up care offered to patients presenting to ED with SH should not be based solely on the result of screening tools. The use of multicentre data helped to maximise the generalisability of the tool, but we need to further verify its external validity in other areas of the UK and in other countries.

 

Bergen, H., Hawton, K., Waters, K., Cooper, J. & Kapur, N. (2010). Psychosocial assessment and repetition of self-harm: The significance of single and multiple repeat episode analyses. Journal of Affective Disorders 127, 257-265. Link to the paper.

Self-harm is a common reason for presentation to the Emergency Department. An important question is whether psychosocial assessment following self-harm affects further repetition of self-harm. In this study we found that psychosocial assessment reduced the risk of a single repeat episode by 51% in individuals without a history of psychiatric treatment, and by 26% in individuals with a history of psychiatric treatment. For recurrent repetition of up to 5 episodes, psychosocial assessment decreased risk of a further episode by 13%. Appropriate statistical models which take account of multiple episodes by the same person are needed to accurately assess risk in recurrent repetition. 
Conclusions: Psychosocial assessment is beneficial in reducing repeated self-harm in people both with and without a history of psychiatric care. Our finding supports the recommendation from the National Institute for Clinical Excellence that all individuals presenting to the emergency department with self-harm should receive a psychosocial assessment.

 

Lilley, R., Owens, D., Horrocks, J., House, A., Noble, R., Bergen, H., Hawton, K., Casey, D., Simkin, S., Murphy, E., Cooper, J. & Kapur, N. (2008). Hospital care and repetition following self-harm: a multicentre comparison of self-poisoning and self-injury. British Journal of Psychiatry 192, 440-445. Link to the paper. 

Quantitative research about self-harm largely deals with self-poisoning, despite the high incidence of self-injury. We compared patterns of hospital care and repetition associated with self-poisoning and self-injury. Demographic and clinical data were collected in a multicentre, prospective cohort study, involving 10,498 consecutive episodes of self-harm at six English teaching hospitals. Compared with those who self-poisoned, people who cut themselves were more likely to have self-harmed previously and to have received support from mental health services, but they were far less likely to be admitted to the general hospital or receive a psychosocial assessment. Although only 17% of people repeated self-harm during the 18-month study, survival analysis that takes account of all episodes revealed a repetition rate of 33% in the year following an episode. There was a significantly greater rate of repetition after episodes of self-cutting (47%) and than after self-poisoning (31%). Of those who repeated, a third switched method of self-harm.
Conclusions: Hospital services offer less to people who have cut themselves, although they are far more likely to repeat, than to those who have self-poisoned. Attendance at hospital should result in psychosocial assessment of needs regardless of method of self-harm.

 

Kapur, N., Murphy, E., Cooper, J., Bergen, H., Hawton, K., Simkin, S., Casey, D., Horrocks, J., Lilley, R., Noble, R. & Owens, D. (2008). Psychosocial assessment following self-harm: results from the Multi-Centre Monitoring of Self-Harm Project. Journal of Affective Disorders 106, 285-293. Link to the paper.

A good quality assessment is central to the management of self-harm, but not all people are assessed following presentation to hospital. It is unclear whether being assessed affects the risk of repeating self-harm. In this study we used 18 months of data from the Multicentre Study of Self-harm in England (covering Oxford, Leeds and Manchester). We found that 60% of episodes resulted in a specialist psychosocial assessment. Factors which made an assessment more likely included age over 55 years, current psychiatric treatment, admission to a medical ward, and taking an overdose of antidepressants. Factors which made an assessment less likely included unemployment, self-cutting, attending outside normal working hours, and self-discharge. Overall we found no link between assessment and self-harm repetition, but there were differences between hospitals - assessments were protective in one hospital but associated with an increased risk of repetition in another. 
Conclusions: Many people who harm themselves, including potentially vulnerable individuals, do not receive an adequate assessment while at hospital. Staff should be aware of this. Identifying which components of psychosocial assessment are particularly helpful will help to shape the way we manage self-harm in the future.

Ness, J., Hawton, K., Bergen, H., Waters, K., Kapur, N., Cooper, J., Steeg, S., Clarke, M. (2016). High volume repeaters of self-harm: Characteristics, patterns of emergency department attendance and subsequent deaths based on findings from the Multicentre Study of Self-harm in England. CRISIS, Oct 12:1-11. Link to the paper.

Self-harm is a behaviour which is often repeated and is associated with an increased risk of dying by suicide. In this study, we explored how common repeat attendance to the emergency department (ED) following self-harm was and the patterns of attendance amongst those who attended most frequently. Data collected as part of the Multicentre Study of Self-harm in England. High volume repetition was defined as ≥15 attendances to the ED following self-harm within four years. Every person with high volume repetition had an ED attendance timeline created. These timelines were then subjected to an executive sorting task and a hierarchical cluster analysis to try to identify similar groupings of patterns of presentations. We found that a very small number of people attending ED following self-harm did so frequently. Thus just 0.6% of patients fitted our high volume repeater definition, but they accounted10% of all the self-harm attendances to ED. Three types of attendance patterns were identified: 1. intermittent attendance with few clusters, 2. intermittent attendance with multiple clusters, 3. most frequent attendees. We also found that a greater proportion of those attending ED frequently subsequently died from external causes (e.g. accidental, suicide) compared to those who did not repeat self-harm frequently.
Conclusions: Individuals who frequently attend ED following self-harm represent a very small proportion of self-harm patients but account for a large number of all self-harm attendances. The need for early intervention is highlighted by the large clustered nature of attendances and the higher frequency of death from external causes. The research methods used in this study offer a new way of exploring very frequent repeat self-harm behaviour, which could have both clinical and research benefits.

 

Turnbull, P., Webb, R., Kapur, N., Clements, C., Bergen, H., Hawton, K., Ness, J., Waters, K., Townsend, E., Cooper, J. (2015). Variation by ethnic group in premature mortality risk following self-harm: a multicentre cohort study in England. BMC Psychiatry. Link to the paper.

People who self-harm are known to be at increased risk of dying early. There are differences in rates of self-harm and risk factors associated with self-harm for different ethnic groups, yet little is known about the influence of ethnicity on death after self-harm. We used data from the Multicentre Study of Self-harm in England, linked to a national mortality dataset, to investigate early death in South Asian and Black people, and compared them to early deaths in White people following a hospital presentation for self-harm. After adjusting for age, gender and area-level socioeconomic deprivation, we found that the risk of death by any cause was lower in South Asian and Black people after self-harm, than it was in White people after self-harm. Risk of dying by suicide after self-harm was much lower in Black people than in White people. We also found that many risk factors usually associated with dying early after self-harm, such as psychiatric treatment or concurrent alcohol misuse, were seen less often in South Asian and Black people than in White people. 
Conclusions: The risk of dying after a hospital presentation for self-harm is lower in South Asian and Black people than White people in the UK. South Asian and Black people also have fewer risk factors for premature death. This may truly reflect a lower risk of dying early after a hospital presentation for self-harm, but may also be a result of different help-seeking strategies employed by South Asian and Black people, who may be less likely that White people to attend hospital following self-harm. 

 

Kapur, N., Steeg S, Turnbull, P., Webb, R., Bergen, H., Hawton, K., Geulayov, G., Townsend, E., Ness, J., Waters, K., Cooper, J. (2015). Hospital management of suicidal behavious and subsequent mortality: a prospective cohort study. Lancet Psychiatry. Link to the paper.

Self-poisoning and self-injury are associated with a high risk of suicide or death from any cause but the effect of routine aspects of hospital management on mortality risk is unknown. We followed up 38,145 people admitted to hospital following self-harm between 2000 and 2010 to investigate the association between the treatments patients receive in hospital (psychosocial assessment, medical admission, psychiatric admission, referral for mental health follow-up) and their subsequent risk of death. We found that most aspects of management were associated with increased mortality risk, and psychiatric admission or mental health follow-up were associated with the highest risk of death. This finding suggests that hospitals might be reserving the most intensive treatments for the patients at greatest need. There were significant interactions by subgroup, suggesting that the association between management and outcome was not the same for all people. In particular, admission to an inpatient psychiatric ward might reduce 12 month all-cause mortality in some high risk groups (men, adults older than 65 years, and those with a history of self-harm). 
Conclusions:The findings of this study show how high the risk of suicide and other premature death can be for patients who attend A&E after self-harm. The risk is elevated further for people requiring the most intensive forms of clinical management, such as referral to outpatient mental health services and psychiatric hospital admission. Reassuringly, these findings indicate clinicians are appropriately allocating treatment. When we take into account some of the differences between patients, admission to inpatient psychiatry appears to have benefits in terms of reducing suicide and other early death in some subgroups. 

 

Hawton, K., Bergen, H., Cooper, J., Turnbull, P., Waters, K., Ness, J. & Kapur, N. (2015) Suicide following self-harm: findings from the Multicentre Study of self-harm in England, 2000-2012. Journal of Affective Disorders 175, 147-51. Link to the paper.

Self-harm is a key risk factor for suicide and it is important to have up-to-date information on the extent of this risk. We followed up 40,346 self-harm patients identified in the three centres of the Multicentre Study of Self-harm in England between 2000 and 2010. By the end of 2012, 2704 individuals had died. Nearly one in five of the deaths were by suicide (including open verdicts), which occurred in 1.6% of patients (2.6% of males and 0.9% of females). Overall, 0.5% of individuals died by suicide in the first year (including 0.82% of males and 0.27% of females), during which time the risk was 49 times greater than the risk of suicide in the general population. Risk of suicide increased with age. While self-poisoning had been the most frequent method of self-harm, hanging was the most common method of subsequent suicide, particularly in males. The number of suicides was probably a considerable underestimate as there were also a large number of deaths recorded as accidents, the majority of which were poisonings (these often involving psychotropic drugs).
Conclusions: The findings underline the importance of suicide prevention initiatives focused on the self-harm population, especially during the initial months following an episode of self-harm. Estimates of using suicide and open verdicts may underestimate the true risk of suicide following self-harm and inclusion of accidental poisonings may be warranted in future risk estimates. 

 

Owens, D, Kelley, R., Munyombwe, T., Bergen, H., Hawton, K., Cooper, J., Ness, J., Waters, K., West, R. & Kapur N. (2015). Switching methods of self-harm at repeat episodes: Findings from a multicentre cohort study. Journal of Affective Disorders 180, 44-51. Link to the paper.

It is recognised that people who self-harm more than once may switch from one method of self-harm to another. In this study we aimed to find out the frequency, pattern, determinants and characteristics of changes in methods of self-harm in individuals presenting to general hospitals. We used information on over 33,000 consecutive self-harm episodes identified in six general hospitals in Manchester, Derby and Oxford between 2003 and 2007. Over an average of 30 months of follow-up, 23% of people repeated self-harm and one-third of them switched method, often with a very short time interval between episodes, and especially where the person was male, younger, or had self-harmed previously. Self-poisoning was far less likely than other methods to be followed by switching of method. 
Conclusions: When self-harm is repeated the method used often changes, but the nature of the change may be relatively unpredictable, except that this less often occurs following self-poisoning. Clinicians should therefore avoid potentially misleading assumptions about risks or needs of patients based simply on the method of harm. 

 

Bergen, H., Hawton, K., Webb, R., Cooper, J., Steeg, S., Haigh, M., Ness, J., Waters, K. & Kapur N. (2014). Alcohol-related mortality following self-harm: a multicentre cohort study. JRSM Open, 5(8), 2054270414533326. Link to the paper.

We assessed alcohol-related premature death in people who self-harm compared to the general population, including differences in risk according to socioeconomic status. We studied 39,014 individuals aged 15 years or more who presented to emergency departments in Oxford, Manchester or Derby following self-harm between 2000 and 2010 and identified deaths up to the end of 2012. After an average (median) of 7.5 years' follow-up, 2695 individuals (6.9%) had died, including significantly more males (9.5%) than females (5.0%). Overall, 307 (11.4%) individuals had died from alcohol-related causes. Alcohol-related death was far more frequent than expected in both males and females. Taking account of the age when deaths occurred, deaths from alcohol-related causes occurred some 34 years earlier than would have been expected in male patients and some 38 years earlier in female patients. This was unrelated to the level of socioeconomic deprivation in the areas where individuals lived. Alcohol-related death was associated with unemployed/sick/disabled status, alcohol use during self-harm, referral to drug/alcohol services and lack of psychosocial assessment following self-harm.
Conclusions: Alcohol-related deaths contribute substantially to premature deaths in self-harm patients. During the assessment of patients following self-harm there should be a particular focus on possible alcohol misuse to enable early detection and treatment of alcohol problems. This requires easy access to alcohol misuse services. It could contribute to prevention of premature death from alcohol-related causes.

 

Steeg, S., & Kapur, N. (2011). How do methods used in previous episodes of non-fatal self-harm relate to eventual suicide? Journal of Affective Disorders, 136, 526-533. Link to the paper.

Methods used at an index episode of non-fatal self-harm may predict risk of future suicide. In this study we investigated suicide risk associated with most recent non-fatal method, and whether or not change in method was important. Increased risks of suicide by injury were found for individuals who used self-cutting, hanging/ asphyxiation, carbon monoxide/other gas, traffic-related and other self-injury at the last episode of self-harm (1.8 to 5-fold risks compared to poisoning). One-third who died by suicide used the same method for their last self-harm and for suicide, including 41% who self-poisoned.
Conclusions: Individuals using dangerous methods of self-harm, such as hanging and inhalation of car exhaust or other gas, should receive particularly intensive follow-up after a non-fatal attempt. There appears to be no pattern in change of method in the last two episodes of self-harm which leads to increased risk of suicide. Other factors need to be considered, as well as last method non-fatal self-harm, when suicide risk is assessed in the emergency department. 

See also: Evid Based Mental Health

 

Bergen, H., Hawton, K., Kapur, N., Cooper, J., Steeg, S., Ness, J., & Waters, K. (2011). Shared characteristics of suicides and other unnatural deaths following non-fatal self-harm? A multicentre study of risk factors. Psychological Medicine. Link to the paper.

In this study we compared the risk of death by suicide and by accidental cause in individuals who had previously presented to Emergency Departments with self-harm during 2000 to 2007. Individuals were followed up to the end of 2009 using national death registers to identifydeaths and ascertain the cause. There were many risks common to both types of death, such as being male, of older age and having a history of psychiatric treatment. Individuals who died by suicide were more likely to have previously self-harmed, particularly using violent methods or self-cutting rather than poisoning, and to have mental health problems. Individuals dying by accidental narcotic poisoning (or non-narcotic poisoning) were likely to have had recreational/ illicit drug (or alcohol) problems at their last episode of non-fatal self-harm.
Conclusions: Individuals who died by suicide and by accidents have many things in common, such as unemployment or disability, life problems, alcohol or drug use and mental illness. These difficulties may lead to a variety of self-destructive behaviour. Differences seem to be partly related to the criteria coroners use in reaching verdicts. Drug or alcohol use may make an accidental verdict more likely than a suicide verdict. 

See also Evid Based Mental Health 2012;15:101

 

Bergen, H., Hawton, K., Waters, K., Cooper, J. & Kapur, N. (2010).Psychosocial assessment and repetition of self-harm: The significance of single and multiple repeat episode analyses. Journal of Affective Disorders 123, 95-101. Link to the paper.

Self-harm is a common reason for presentation to the Emergency Department. An important question is whether psychosocial assessment following self-harm affects further repetition of self-harm. In this study we found that psychosocial assessment reduced the risk of a single repeat episode by 51% in individuals without a history of psychiatric treatment, and by 26% in individuals with a history of psychiatric treatment. For recurrent repetition of up to 5 episodes, psychosocial assessment decreased risk of a further episode by 13%. Appropriate statistical models which take account of multiple episodes by the same person are needed to accurately assess risk in recurrent repetition.
Conclusions: Psychosocial assessment is beneficial in reducing repeated self-harm in people both with and without a history of psychiatric care. Our finding supports the recommendation from the National Institute for Clinical Excellence that all individuals presenting to the emergency department with self-harm should receive a psychosocial assessment.

Hawton, K., Ferrey, A., Casey, D., Wells, C., Fuller, A., Bankhead, C., Clements, C., Ness, J., Gunnell, D., Kapur, N., Geulayov, G. (2019). Relative toxicity of analgesics commonly used for intentional self-poisoning: A study of case fatality based on fatal and non-fatal overdoses. Journal of Affective Disorders, 246:814-819.

Ferrey, A.E., Geulayov, G., Casey, D., Wells, C., Fuller, A., Bankhead, C., Ness, J., Clements, C., Gunnell, D., Kapur, N., Hawton, K. (2018). Relative toxicity of mood stabilisers and antipsychotics: case fatality and fatal toxicity associated with self-poisoning. BMC Psychiatry, 18:399.

Geulayov, G., Ferrey, A., Casey, D., Wells, C., Fuller, A., Bankhead, C., Gunnell, D., Clements, C., Kapur, N., Ness, J., Waters, K., Hawton, K. (2018). Relative toxicity of benzodiazepines and hypnotics commonly used for self-poisoning: An epidemiological study of fatal toxicity and case fatality. Journal of Psychopharmacology, 32:654-662.

Hawton, K., Bergen, H., Simkin, S., Arensman, E., Corcoran, P., Cooper, J., Waters, K., Gunnell, D., and Kapur, N. (2011). Impact of different pack sizes of paracetamol in the United Kingdom and Ireland on intentional overdoses: a comparative study. BMC Public Health, 11:460.  Link to the paper.

In order to reduce fatal self-poisoning legislation was introduced in the UK in 1998 to restrict pack sizes of paracetamol sold in pharmacies (maximum 32 tablets) and non-pharmacy outlets (maximum 16 tablets), and in Ireland in 2001, but with smaller maximum pack sizes (24 and 12 tablets). In order to determine whether this resulted in smaller overdoses of paracetamol in Ireland compared with the UK, we used data on general hospital presentations for non-fatal self-harm for 2002 – 2007 from the Multicentre Study of Self-harm in England (six hospitals) and from the National Registry of Deliberate Self-harm in Ireland to compare sizes of overdoses of paracetamol in the two settings. We found clear peaks in numbers of non-fatal overdoses, associated with the different maximum pack sizes of paracetamol in pharmacy and non-pharmacy outlets in both England and Ireland. Significantly more pack equivalents (based on maximum non-pharmacy pack sizes) were used in (based on maximum non-pharmacy pack sizes) were used in overdoses in Ireland compared with England. However, the overall size of overdoses did not differ significantly between England and Ireland.
Conclusions: The difference in paracetamol pack size legislation between England and Ireland does not appear to have resulted in a major difference in sizes of overdoses. This is because more pack equivalents are taken in overdoses in Ireland, possibly reflecting differing enforcement of sales advice. Differences in access to clinical services may also be relevant. 

Hawton, K., Bergen, H., Waters, K., Murphy, E., Cooper, J. & Kapur, N. (2011). Impact of withdrawal of the analgesic co-proxamol in the UK on non-fatal self-poisoning. Crisis 32, 81-87. Link to the paper.

In early 2005 the UK Committee on Safety of Medicines (CSM) announced gradual withdrawal of the analgesic co-proxamol because of its adverse benefit/safety ratio, especially its extensive use for intentional and accidental fatal poisoning. Prescriptions of co-proxamol were reduced in the 3-year withdrawal phase (2005 to 2007) following the CSM announcement. We assessed the impact of the CSM announcement in January 2005 to withdraw co-proxamol on non-fatal self-poisoning with co-proxamol and other analgesics. We used data on general hospital presentations for non-fatal self-poisoning (five hospitals in three centers in England). When the 3-year withdrawal period 2005-2007 was compared with 2000-2004, we found a marked reduction in the number of episodes of non-fatal self-poisoning episodes involving co-proxamol following the CSM announcement (an estimated 62% reduction over the period 2005 to 2007 compared to 2000 to 2004). There was no evidence of an increase in non-fatal self-poisoning episodes involving other analgesics (co-codamol, codeine, co-dydramol, dihydrocodeine, and tramadol) in relation to the CSM announcement over the same period, nor a change in the number of all episodes of self-poisoning.
Conclusions: The withdrawal of co-proxamol in the UK appears to have resulted in reduced non-fatal self-poisoning with co-proxamol, without significant substitution with other analgesics. This finding is in keeping with that found for suicide.

Hawton, K., Bergen, H., Simkin, S., Cooper, J., Waters, K., Gunnell, D. & Kapur, N. (2010). Toxicity of antidepressants: study of rates of suicide relative to prescribing and non-fatal overdose. British Journal of Psychiatry 196, 354-358.

Self-poisoning is a common method of suicide and often involves ingestion of antidepressants. Information on the relative toxicity of antidepressants is therefore extremely important. We assessed the relative toxicity of seven specific antidepressants, by comparing numbers of prescriptions for each antidepressant with poisoning suicide deaths in England and Wales involving each antidepressant (alone), and non-fatal self-poisoning episodes with each antidepressant presenting to six general hospitals (in Oxford, Manchester and Derby), between 2000 and 2006. We calculated fatal toxicity for each antidepressant , based on ratio of rates of deaths to prescriptions, and case fatality, based on ratio of rates of deaths to non-fatal self-poisonings. Fatal toxicity and case fatality indices provided very similar results. Case fatality rate ratios showed far greater toxicity for tricyclic antidepressants than the SNRI antidepressant venlafaxine and the NaSSA antidepressant mirtazapine, both of which had approximately four times greater toxicity than the SSRIs. Within the tricyclic antidepressants, compared with amitriptyline both dosulepin and doxepin were considerably more toxic. Within the SSRIs, citalopram had a higher case fatality than the other SSRIs.
Conclusions: There are wide differences in toxicity not only between classes of antidepressants, but also within classes. The findings are relevant to prescribing decisions, especially in individuals at risk, in whom clinicians should use less toxic antidepressants wherever possible. They are also relevant to regulatory authority policy regarding advice to clinicians. The findings have been communicated to the UK Medicines and Healthcare Products Regulatory Agency. 

Bergen, H., Murphy, E., Cooper, J., Kapur, N., Stalker, C., Waters, K. & Hawton, K. (2010). A comparative study of non-fatal self-poisoning with antidepressants relative to prescribing in three centres in England. Journal of Affective Disorders 123, 95-101. Link to the paper.

Antidepressants are used frequently in non-fatal self-poisoning. There are national guidelines for prescribing antidepressants. In this study we compared the use of specific antidepressants (amitriptyline and dosulepin (tricyclics), citalopram, fluoxetine, paroxetine and sertraline (selective serotonin reuptake inhibitors) and venlafaxine (serotonin norepinephrine reuptake inhibitor)) for non-fatal self-poisoning relative to prescribing, in three centres in England, from 2004 to 2006. We found marked variation between centres in the ratio of rates of self-poisoning to prescribing for specific antidepressants. This is likely due to differences in both prescribing practices (despite clear national guidance) and patient characteristics. 
Conclusions: Clinicians should consider carefully the toxicity, and the risk of overdose, when prescribing antidepressants for their patients. 

Bergen, H., Hawton, K., Murphy, E., Cooper, J., Kapur, N., Stalker, C. & Waters, K. (2009). Trends in prescribing and self-poisoning in relation to UK regulatory authority warnings against use of SSRI antidepressants in under-18 year-olds. British Journal of Clinical Pharmacology 68, 618-629.  Link to the paper.

Following the UK Medicines and Healthcare products Regulatory Authority (MHRA) warning in December 2003 not to prescribe selective serotonin reuptake inhibitor (SSRI) antidepressants, except fluoxetine, to under 18 year-olds, the prescribing of antidepressants declined in this group. Other studies had found no related changes in rates of suicide or hospital admissions for self-harm. In this study we found that presentations to general hospitals for non-fatal self-poisoning with SSRI antidepressants (but not fluoxetine) declined in 12-19 year-olds in three centres in England in line with UK prescribing trends. There was some evidence of a possible small substitution effect from use of other SSRIs for non-fatal self-poisoning to use of fluoxetine. Conclusions: The MHRA warning in 2003 not to prescribe SSRI antidepressants to young people was effective, as prescriptions for these drugs decreased after this date. Importantly, the occurrence of non-fatal self-harm in 12-19 year-olds in the three centres in this study was stable before and after 2003, with no major change of method or increase in self-injury. Thus the reduction in prescribing of SSRIs did not appear to result in increased self-harm in young people.