Materiały źródłowe

• #1 Predicting self-harm within six months after initial presentation to youth mental health services: A machine learning study

  https://pmc.ncbi.nlm.nih.gov/articles/PMC7775066/

  A priority for health services is to reduce self-harm in young people. Predicting self-harm is challenging due to their rarity and complexity, however this does not preclude the utility of prediction models to improve decision-making regarding a service response in terms of more detailed assessments and/or intervention. The aim of this study was to predict self-harm within six-months after initial presentation. […] Prediction models for self-harm may have utility to identify a large sub population who would benefit from further assessment and targeted (low intensity) interventions. Such models could enhance health service approaches to identify and reduce self-harm, a considerable source of distress, morbidity, ongoing health care utilisation and mortality. […] The decision curve analysis indicates that there was a net benefit of these models over a treat everybody approach, suggesting the potential to allocate targeted assessments and interventions in addition to those broad health service strategies.

• #1 Self-harm in young adolescents (12–16 years): onset and short-term continuation in a community sample | BMC Psychiatry | Full Text

  https://bmcpsychiatry.biomedcentral.com/articles/10.1186/1471-244X-13-328

  Self harm in young adolescents is common with one in four reporting self-harming thoughts and one in six engaging in self-harming behaviour over a one year period. […] Self-harm is already established by 12/13 years of age and for over half of our sample, self-harming thoughts and behaviour persisted over the year. […] The continuation of self-harming thoughts/behaviours and the relationship with mood, alcohol, cannabis and drug use, bullying and school connectedness is summarised in Table 4. […] Self-harming thoughts and behaviour amongst young adolescents are evident by school year 8 (aged 12-13), with girls in school year 9 (aged 13-14) being at particular risk. […] These thoughts and behaviours can persist over time, with 60% of those reporting self-harming thoughts and 55% of those reporting self-harming acts doing so across both assessment periods.

• #1 Predicting suicide following self-harm: systematic review of risk factors and risk scales | The British Journal of Psychiatry | Cambridge Core

  https://www.cambridge.org/core/journals/the-british-journal-of-psychiatry/article/predicting-suicide-following-selfharm-systematic-review-of-risk-factors-and-risk-scales/C9D595168EDF06401A823E2E968915E1

  People with a history of self-harm are at a far greater risk of suicide than the general population. […] The four risk factors that emerged, although of interest, are unlikely to be of much practical use because they are comparatively common in clinical populations. […] The use of these scales, or an over-reliance on the identification of risk factors in clinical practice, may provide false reassurance and is, therefore, potentially dangerous. […] Comprehensive psychosocial assessments of the risks and needs that are specific to the individual should be central to the management of people who have self-harmed. […] There is robust pooled evidence from 12 studies to show that four factors (previous episodes of self-harm, suicidal intent, poor physical health and male gender) are associated with a higher risk of dying by suicide following the index episode.

• #1 Predicting future self-harm or suicide in adolescents: a systematic review of risk assessment scales/tools

  https://pmc.ncbi.nlm.nih.gov/articles/PMC6731844/

  This systematic review aimed to evaluate the ability of risk tools to predict the future episodes of suicide/self-harm in adolescents. […] The ability of the tools to correctly identify those adolescents going on to make a self-harm/suicide attempt ranged from 27% (95% CI 10.7% to 50.2%) to 95.8% (95% CI 78.9% to 99.9%). […] The predictive ability of these tools varies greatly. No single tool is suitable for predicting a higher risk of suicide or self-harm in adolescent populations. […] The predictive validity of the Self-Harm Questionnaire (SHQ) was reported at 3 months. This was: sensitivity 94.7%, specificity 34.6%, PPV 25.4% and NPV 96.6%. This tool performed well at identifying high-risk patients, but performed poorly at identifying low-risk patients. […] The predictive ability of the Ask Suicide Screening Questions (ASQ) was reported at 6 months as: sensitivity 95.8%, specificity 5.8%, PPV 16.8% and NPV 87.5%.

• #1 Predicting self-harm within six months after initial presentation to youth mental health services: A machine learning study

  https://pmc.ncbi.nlm.nih.gov/articles/PMC7775066/

  A better understanding of key model predictors may be helpful to inform clinical decision making for reducing self-harm. […] The present work supports the view that data driven, and machine learning methods have the potential to advance clinical decision making for self-harm. This study demonstrates the potential clinical utility of prediction models to identify a large sub population who may benefit from targeted (low intensity) interventions in addition to the broad health service prevention strategies. Enhancing how health services identify and respond to self-harm is a critical priority, not simply because of the risk they confer for future suicide, but due to the significant distress, morbidity and ongoing health care utilisation associated with self-harm.

• #1 Predicting suicide following self-harm: systematic review of risk factors and risk scales | The British Journal of Psychiatry | Cambridge Core

  https://www.cambridge.org/core/journals/the-british-journal-of-psychiatry/article/predicting-suicide-following-selfharm-systematic-review-of-risk-factors-and-risk-scales/C9D595168EDF06401A823E2E968915E1

  However, when assessing people following an act of self-harm, being able to identify these associated factors is still unlikely to help us to predict the risk of later suicide, because these characteristics are common in clinical populations. […] The idea of risk assessment as risk prediction is a fallacy and should be recognised as such. […] We are simply unable to say with any certainty who will and will not go on to have poor outcomes. […] It is also important to recognise that different studies used different risk scales, and some used different cut-off scores for the same risk scales. […] Taking these limitations into account, we can conclude that there is insufficient evidence to support the use of risk scales and tools in clinical practice.

• #1 Risk of death by suicide following self-harm presentations to healthcare: development and validation of a multivariable clinical prediction rule (OxSATS) | BMJ Mental Health

  https://mentalhealth.bmj.com/content/26/1/e300673

  An 11-item risk model to predict suicide was developed using sociodemographic and clinical risk factors, and showed good discrimination and calibration in external validation. […] OxSATS accurately predicts 12-month risk of suicide. […] Using a clinical prediction score may assist clinical decision-making and resource allocation. […] The study underscores the importance of considering probability scores for risk prediction, which are widely used in prognostic tools in cardiovascular and cancer medicine. […] Without linkage to interventions, implementing a risk prediction model on its own will not improve outcomes. […] Future work will need to consider how the tool can be used, at what point, and how it can be linked to treatment. […] From a clinical perspective, one strength of a prediction model is that it can improve consistency, especially in busy clinical settings and where assessment is conducted by people with different professional and training backgrounds, anchor decisions in empirical evidence, highlight the role of certain modifiable factors, and provide an opportunity to transparently discuss risk with patients and their carers. […] The tool is based on 11 predictors, and has been translated into a simple online tool with probability scores for suicide at 6 and 12 months after a self-harm presentation.

• #1 Self-harm in young adolescents (12–16 years): onset and short-term continuation in a community sample | BMC Psychiatry | Full Text

  https://bmcpsychiatry.biomedcentral.com/articles/10.1186/1471-244X-13-328

  Our results indicate that thoughts and acts of self-harm are evident in those aged 14 years. […] This suggests the need for self-harm prevention programmes to be more focused upon younger adolescents (aged 12-13) before self-harming has become established. […] Self-harm behaviour at follow up was more likely to be preceded, and then accompanied, by self-harm thoughts. […] This suggests that focusing upon improving general mental health and developing positive cognitive skills is likely to have an impact on numbers engaging in self-harm behaviours. […] Our findings indicate that self-harm in young adolescents is common and can be persistent, yet few specifically seek help for psychological problems. […] This suggests a need to raise awareness of self-harm and potential risk factors such as low mood, bullying and drug misuse amongst those who have regular contact with young adolescents in the community, schools and primary care.

• #2 Predicting suicide following self-harm: systematic review of risk factors and risk scales | The British Journal of Psychiatry | Cambridge Core

  https://www.cambridge.org/core/journals/the-british-journal-of-psychiatry/article/predicting-suicide-following-selfharm-systematic-review-of-risk-factors-and-risk-scales/C9D595168EDF06401A823E2E968915E1

  People with a history of self-harm are at a far greater risk of suicide than the general population. […] The four risk factors that emerged, although of interest, are unlikely to be of much practical use because they are comparatively common in clinical populations. […] The use of these scales, or an over-reliance on the identification of risk factors in clinical practice, may provide false reassurance and is, therefore, potentially dangerous. […] Comprehensive psychosocial assessments of the risks and needs that are specific to the individual should be central to the management of people who have self-harmed. […] There is robust pooled evidence from 12 studies to show that four factors (previous episodes of self-harm, suicidal intent, poor physical health and male gender) are associated with a higher risk of dying by suicide following the index episode.

• #2

  https://link.springer.com/article/10.1007/s00127-022-02415-7

  The proportion of young people with DSH or SA at baseline, who repeated DSH or SA in 12 months of follow-up was high (53.5%,116/217). […] DSH via cutting or burning at baseline was associated with the greatest odds of either DSH or SA during follow-up. […] Our findings suggest risk factors within clinical populations of young people are different to those of population samples of young people. […] Our findings of the importance of clinical symptoms to future self-harm or suicidal behaviour suggest that leading theories of suicidal behaviour may require adaption to clinical settings. […] Both models performed poorly and were only slightly better at predicting outcome than chance. […] Given the significant role clinical symptoms play in predicting deliberate self-harm or suicide attempts in young people accessing clinical services, the key suicide prevention in this cohort should be the delivery of high-quality clinical care that aims for recovery, of which amelioration of symptoms is one component.

• #2 Scales for predicting risk following self-harm: an observational study in 32 hospitals in England | BMJ Open

  https://bmjopen.bmj.com/content/4/5/e004732

  There is little consensus over the best instruments for risk assessment following self-harm. […] Further research to evaluate the impact of scales following an episode of self-harm is warranted using prospective designs. Until then, it is likely that the indiscriminant use of risk scales in clinical services will continue. […] Hospitals which used published scales as a component of their risk assessments had a lower median rate of repeat self-harm at 6months than hospitals which did not. […] When we adjusted our models for possible confounders, the apparent protective effect of the use of scales was attenuated. […] Interestingly, our study suggested that services which used risk scales may have had lower repetition rates than services which did not. […] A future study including follow-up of self-harm in the community might help determine the influence of this on subsequent help-seeking behaviour.

• #2 Predicting future self-harm or suicide in adolescents: a systematic review of risk assessment scales/tools

  https://pmc.ncbi.nlm.nih.gov/articles/PMC6731844/

  This review is the first to explore the use of tools to predict future self-harm/suicide attempts in an adolescent population. It has shown that the current limited amount of primary evidence means at present no individual tool can be identified as performing better than another or is a sufficiently accurate predictor of self-harm/suicide risk.

• #2

  https://link.springer.com/article/10.1007/s00127-022-02415-7

  Machine learning (ML) has shown promise in modelling future self-harm but is yet to be applied to key questions facing clinical services. […] The mean performance of models predicting SA (AUC: 0.82) performed better than the models predicting DSH (AUC: 0.72), with mean positive predictive values (PPV) approximately twice that of the prevalence (SA prevalence 14%, PPV: 0.32, DSH prevalence 22%, PPV: 0.40). […] History of DSH and clinical symptoms of common mental disorders, rather than social and demographic factors, were the most important variables in modelling future behaviour. […] The performance of models predicting outcomes in key sub-cohorts, those with new-onset or repetition of DSH or SA during follow-up, was poor. […] These findings may indicate that the performance of models of future DSH or SA may depend on knowledge of the individuals recent history of either behaviour.

• #2 Predicting self-harm within six months after initial presentation to youth mental health services: A machine learning study

  https://pmc.ncbi.nlm.nih.gov/articles/PMC7775066/

  A better understanding of key model predictors may be helpful to inform clinical decision making for reducing self-harm. […] The present work supports the view that data driven, and machine learning methods have the potential to advance clinical decision making for self-harm. This study demonstrates the potential clinical utility of prediction models to identify a large sub population who may benefit from targeted (low intensity) interventions in addition to the broad health service prevention strategies. Enhancing how health services identify and respond to self-harm is a critical priority, not simply because of the risk they confer for future suicide, but due to the significant distress, morbidity and ongoing health care utilisation associated with self-harm.

• #2 Predicting suicide following self-harm: systematic review of risk factors and risk scales | The British Journal of Psychiatry | Cambridge Core

  https://www.cambridge.org/core/journals/the-british-journal-of-psychiatry/article/predicting-suicide-following-selfharm-systematic-review-of-risk-factors-and-risk-scales/C9D595168EDF06401A823E2E968915E1

  However, when assessing people following an act of self-harm, being able to identify these associated factors is still unlikely to help us to predict the risk of later suicide, because these characteristics are common in clinical populations. […] The idea of risk assessment as risk prediction is a fallacy and should be recognised as such. […] We are simply unable to say with any certainty who will and will not go on to have poor outcomes. […] It is also important to recognise that different studies used different risk scales, and some used different cut-off scores for the same risk scales. […] Taking these limitations into account, we can conclude that there is insufficient evidence to support the use of risk scales and tools in clinical practice.

• #2 Risk of death by suicide following self-harm presentations to healthcare: development and validation of a multivariable clinical prediction rule (OxSATS) | BMJ Mental Health

  https://mentalhealth.bmj.com/content/26/1/e300673

  An 11-item risk model to predict suicide was developed using sociodemographic and clinical risk factors, and showed good discrimination and calibration in external validation. […] OxSATS accurately predicts 12-month risk of suicide. […] Using a clinical prediction score may assist clinical decision-making and resource allocation. […] The study underscores the importance of considering probability scores for risk prediction, which are widely used in prognostic tools in cardiovascular and cancer medicine. […] Without linkage to interventions, implementing a risk prediction model on its own will not improve outcomes. […] Future work will need to consider how the tool can be used, at what point, and how it can be linked to treatment. […] From a clinical perspective, one strength of a prediction model is that it can improve consistency, especially in busy clinical settings and where assessment is conducted by people with different professional and training backgrounds, anchor decisions in empirical evidence, highlight the role of certain modifiable factors, and provide an opportunity to transparently discuss risk with patients and their carers. […] The tool is based on 11 predictors, and has been translated into a simple online tool with probability scores for suicide at 6 and 12 months after a self-harm presentation.

• #3 Comparison between clinician and machine learning prediction in a randomized controlled trial for nonsuicidal self-injury | BMC Psychiatry | Full Text

  https://bmcpsychiatry.biomedcentral.com/articles/10.1186/s12888-024-06391-x

  This study explored the predictive abilities of clinicians and a random forest model for treatment response following an internet-delivered emotion regulation treatment for adolescents with NSSI. Findings suggest that both clinicians and the ML algorithm performed above chance but at comparable accuracy rates (63% for clinicians and 67% for ML). However, these accuracies do not meet current thresholds for clinical application. […] Entering the trial with high levels of emotion dysregulation emerged as the most important predictor in the ML, information that was not available from the clinician predictions. These findings indicate that a machine learning model may add information beyond clinician prediction and that patients with more severe difficulties with emotion regulation are more likely to abstain from NSSI the month after this brief internet-delivered intervention.

• #3 Predicting suicide following self-harm: systematic review of risk factors and risk scales | The British Journal of Psychiatry | Cambridge Core

  https://www.cambridge.org/core/journals/the-british-journal-of-psychiatry/article/predicting-suicide-following-selfharm-systematic-review-of-risk-factors-and-risk-scales/C9D595168EDF06401A823E2E968915E1

  However, when assessing people following an act of self-harm, being able to identify these associated factors is still unlikely to help us to predict the risk of later suicide, because these characteristics are common in clinical populations. […] The idea of risk assessment as risk prediction is a fallacy and should be recognised as such. […] We are simply unable to say with any certainty who will and will not go on to have poor outcomes. […] It is also important to recognise that different studies used different risk scales, and some used different cut-off scores for the same risk scales. […] Taking these limitations into account, we can conclude that there is insufficient evidence to support the use of risk scales and tools in clinical practice.

• #3 Comparison between clinician and machine learning prediction in a randomized controlled trial for nonsuicidal self-injury | BMC Psychiatry | Full Text

  https://bmcpsychiatry.biomedcentral.com/articles/10.1186/s12888-024-06391-x

  Preliminary findings indicate comparable prediction accuracy between clinicians and a machine-learning algorithm in the psychological treatment of nonsuicidal self-injury in youth. […] Both clinician (accuracy=0.63) and model-based (accuracy=0.67) predictions achieved significantly better accuracy than a model that classified all patients as reaching NSSI remission (accuracy=0.49 [95% CI 0.41 to 0.58]), however there was no statistically significant difference between them. […] The random forest model and clinician predictions did not differ in terms of accuracy (0.67 and 0.63 respectively, McNemar test=0.205, df=1, p=0.65), and adding the clinician prediction to the random forest model did not improve accuracy compared to the random forest model alone (Accuracy=0.65, McNemar test=0.308, df=1, p=0.58).

• #3 Self-harm in young adolescents (12–16 years): onset and short-term continuation in a community sample | BMC Psychiatry | Full Text

  https://bmcpsychiatry.biomedcentral.com/articles/10.1186/1471-244X-13-328

  Our results indicate that thoughts and acts of self-harm are evident in those aged 14 years. […] This suggests the need for self-harm prevention programmes to be more focused upon younger adolescents (aged 12-13) before self-harming has become established. […] Self-harm behaviour at follow up was more likely to be preceded, and then accompanied, by self-harm thoughts. […] This suggests that focusing upon improving general mental health and developing positive cognitive skills is likely to have an impact on numbers engaging in self-harm behaviours. […] Our findings indicate that self-harm in young adolescents is common and can be persistent, yet few specifically seek help for psychological problems. […] This suggests a need to raise awareness of self-harm and potential risk factors such as low mood, bullying and drug misuse amongst those who have regular contact with young adolescents in the community, schools and primary care.

• #3 Risk of death by suicide following self-harm presentations to healthcare: development and validation of a multivariable clinical prediction rule (OxSATS) | BMJ Mental Health

  https://mentalhealth.bmj.com/content/26/1/e300673

  An 11-item risk model to predict suicide was developed using sociodemographic and clinical risk factors, and showed good discrimination and calibration in external validation. […] OxSATS accurately predicts 12-month risk of suicide. […] Using a clinical prediction score may assist clinical decision-making and resource allocation. […] The study underscores the importance of considering probability scores for risk prediction, which are widely used in prognostic tools in cardiovascular and cancer medicine. […] Without linkage to interventions, implementing a risk prediction model on its own will not improve outcomes. […] Future work will need to consider how the tool can be used, at what point, and how it can be linked to treatment. […] From a clinical perspective, one strength of a prediction model is that it can improve consistency, especially in busy clinical settings and where assessment is conducted by people with different professional and training backgrounds, anchor decisions in empirical evidence, highlight the role of certain modifiable factors, and provide an opportunity to transparently discuss risk with patients and their carers. […] The tool is based on 11 predictors, and has been translated into a simple online tool with probability scores for suicide at 6 and 12 months after a self-harm presentation.

• #4

  https://link.springer.com/article/10.1007/s00127-022-02415-7

  The proportion of young people with DSH or SA at baseline, who repeated DSH or SA in 12 months of follow-up was high (53.5%,116/217). […] DSH via cutting or burning at baseline was associated with the greatest odds of either DSH or SA during follow-up. […] Our findings suggest risk factors within clinical populations of young people are different to those of population samples of young people. […] Our findings of the importance of clinical symptoms to future self-harm or suicidal behaviour suggest that leading theories of suicidal behaviour may require adaption to clinical settings. […] Both models performed poorly and were only slightly better at predicting outcome than chance. […] Given the significant role clinical symptoms play in predicting deliberate self-harm or suicide attempts in young people accessing clinical services, the key suicide prevention in this cohort should be the delivery of high-quality clinical care that aims for recovery, of which amelioration of symptoms is one component.

• #4

  https://link.springer.com/article/10.1007/s00127-022-02415-7

  Machine learning (ML) has shown promise in modelling future self-harm but is yet to be applied to key questions facing clinical services. […] The mean performance of models predicting SA (AUC: 0.82) performed better than the models predicting DSH (AUC: 0.72), with mean positive predictive values (PPV) approximately twice that of the prevalence (SA prevalence 14%, PPV: 0.32, DSH prevalence 22%, PPV: 0.40). […] History of DSH and clinical symptoms of common mental disorders, rather than social and demographic factors, were the most important variables in modelling future behaviour. […] The performance of models predicting outcomes in key sub-cohorts, those with new-onset or repetition of DSH or SA during follow-up, was poor. […] These findings may indicate that the performance of models of future DSH or SA may depend on knowledge of the individuals recent history of either behaviour.

• #5

  https://link.springer.com/article/10.1007/s00127-022-02415-7

  Machine learning (ML) has shown promise in modelling future self-harm but is yet to be applied to key questions facing clinical services. […] The mean performance of models predicting SA (AUC: 0.82) performed better than the models predicting DSH (AUC: 0.72), with mean positive predictive values (PPV) approximately twice that of the prevalence (SA prevalence 14%, PPV: 0.32, DSH prevalence 22%, PPV: 0.40). […] History of DSH and clinical symptoms of common mental disorders, rather than social and demographic factors, were the most important variables in modelling future behaviour. […] The performance of models predicting outcomes in key sub-cohorts, those with new-onset or repetition of DSH or SA during follow-up, was poor. […] These findings may indicate that the performance of models of future DSH or SA may depend on knowledge of the individuals recent history of either behaviour.

• #5 Predicting self-harm within six months after initial presentation to youth mental health services: A machine learning study

  https://pmc.ncbi.nlm.nih.gov/articles/PMC7775066/

  A priority for health services is to reduce self-harm in young people. Predicting self-harm is challenging due to their rarity and complexity, however this does not preclude the utility of prediction models to improve decision-making regarding a service response in terms of more detailed assessments and/or intervention. The aim of this study was to predict self-harm within six-months after initial presentation. […] Prediction models for self-harm may have utility to identify a large sub population who would benefit from further assessment and targeted (low intensity) interventions. Such models could enhance health service approaches to identify and reduce self-harm, a considerable source of distress, morbidity, ongoing health care utilisation and mortality. […] The decision curve analysis indicates that there was a net benefit of these models over a treat everybody approach, suggesting the potential to allocate targeted assessments and interventions in addition to those broad health service strategies.

• #6 Self-harm in young adolescents (12–16 years): onset and short-term continuation in a community sample | BMC Psychiatry | Full Text

  https://bmcpsychiatry.biomedcentral.com/articles/10.1186/1471-244X-13-328

  Our results indicate that thoughts and acts of self-harm are evident in those aged 14 years. […] This suggests the need for self-harm prevention programmes to be more focused upon younger adolescents (aged 12-13) before self-harming has become established. […] Self-harm behaviour at follow up was more likely to be preceded, and then accompanied, by self-harm thoughts. […] This suggests that focusing upon improving general mental health and developing positive cognitive skills is likely to have an impact on numbers engaging in self-harm behaviours. […] Our findings indicate that self-harm in young adolescents is common and can be persistent, yet few specifically seek help for psychological problems. […] This suggests a need to raise awareness of self-harm and potential risk factors such as low mood, bullying and drug misuse amongst those who have regular contact with young adolescents in the community, schools and primary care.

• #7

  https://link.springer.com/article/10.1007/s00127-022-02415-7

  The proportion of young people with DSH or SA at baseline, who repeated DSH or SA in 12 months of follow-up was high (53.5%,116/217). […] DSH via cutting or burning at baseline was associated with the greatest odds of either DSH or SA during follow-up. […] Our findings suggest risk factors within clinical populations of young people are different to those of population samples of young people. […] Our findings of the importance of clinical symptoms to future self-harm or suicidal behaviour suggest that leading theories of suicidal behaviour may require adaption to clinical settings. […] Both models performed poorly and were only slightly better at predicting outcome than chance. […] Given the significant role clinical symptoms play in predicting deliberate self-harm or suicide attempts in young people accessing clinical services, the key suicide prevention in this cohort should be the delivery of high-quality clinical care that aims for recovery, of which amelioration of symptoms is one component.

Zobacz więcej