Computer generated transcript

Warning!
The following transcript was generated automatically from the content and has not been checked or corrected manually.

Self-harm and Suicide in Y oung People Professor Dennis Ougrin, QMULDeclarationofInterest:RoyaltiesfromHodderArnold ▪ Identify the general trends in self-harm and suicide ▪ Recognise the risk factors for self-harm in children and adolescents ▪ Analyse the changes in hospital emergency psychiatric and self-harm presentations following Learning the pandemic outbreak Objectives ▪ Examine the mediating and predicting roles of stringency in lockdown policies ▪ Discuss implicationsto the mental health services during COVID-19 and prepare for future pandemic and lockdownSuicide worldwide Desai et al., Science (2019)Suicide worldwide: Gender and age distribution Naghavi (2019)Suicide in Europe • M Eurostat (2020)Suicide in the UK • Suicide rates in all persons, males and females increased in the past year • Male suicide rate in 2019 is the highest since 2000 • Females suicide rate is the highest since 2004 . Source: Office for National Statistics – Suicides in England and Wales: 2019 registrationsSuicidesof10-18yr .inEnglandand Wales (ONS,2020) Age Deaths 10 0 12 0 13 3 14 6 15 19 16 30 17 28 18 51Suicidebyfemales (ONS,2019)Suicidebymales (ONS,2019)Suicidebythedayoftheweek (Hirvikoskietal,2016)D OR7.55(6.04–9.44) OR2.41(1.14–5.11)–withID OR9.40(7.43–11.90)–withoutIDSuicide in the UK: Suicide methods Hanging, strangulation and suffocation (all grouped together) continued to be the most common method of suicide for both gender in England and Wales, followed by poisoning. Source: Office for National Statistics – Suicides in England and Wales: 2019 registrationsSuicide in the UK: Males aged 10-24 years Suicide rate of males aged 10-24 in England and Wales, registered between 1981 and 2019 2018: 440 deaths (8.2 per 100,000) 2019: 442 deaths (8.2 per 100,000) Source: Office for National Statistics – Suicides in England and Wales: 2019 registrationsSuicide in the UK: Females aged 10-24 years Suicide rate of females aged 10-24 in England and Wales, registered between 1981 and 2019 • In 2019, 159 deaths were recorded (3.1 per 100,000) – the highest recorded rate since 1981 • Suicide rate in females aged 10 to 24 years in England and Wales has increased continuously since 2012 Source: Office for National Statistics – Suicides in England and Wales: 2019 registrationsSelfHarmPrevalenceofselfharminschoolpupilsincountriesparticipatingintheChildandAdolescentself harminEurope(CASE)studybygender(Hawtonetal2006) country self harm meeting study criteria previous year (%) lifetime (%) females males females males England 10.8 3.3 16.9 4.9 Ireland 9.1 2.7 13.5 4.9 The Netherlands 3.7 1.7 5.9 2.5 Belgium 10.4 4.4 15.6 6.8 Norway 10.8 2.5 15.3 4.3 Hungary 5.9 1.7 10.1 3.2 Australia 11.8 1.8 17.1 3.3Self-harminyoungpeoplewith ASD (Duerdenetal,2012) Lifetimeprevalence50% Keypredictors: Abnormalsensoryprocessing Sameness Impairedcognitiveability SocialfunctioningSelf-harm in the UK :In primary care Carr et al. (2016)Self-harm in children and young people Top 5 estimated causesof death in female (top) Rates of self-harm 2007–2016 and male (bottom) adolescents worldwide WHO (2016) Griffin et al. (2018)Self-harm: treatment Largest effect sizes: dialectical behaviour therapy (DBT), cognitive-behavioural therapy (CBT), and mentalization-based therapy (MBT) Ougrin et al. (2015)Self-harm: treatment • Current interventions are overall effective in treating self-harm in adolescence. (d = 0.13, 95% CI 0.04–0.22, p = .004) • DBT-A showed moderate effects in reducing self-harm. (d = 0.51, 95% CI 0.18–0.85, p = .002) Kothgassner et al. (2020)Self-harm: treatment • Treatments for suicidal ideation in adolescence are effective (d = 0.31, 95% CI 0.12–0.50, p = .001) • DBT-A and Family-centred therapy have moderate effects in reducing suicidal ideation Kothgassner et al. (2020)Risk factors for self-harm • Suicidal ideation and depressive symptomatology (Vitieloet al., 2009) • Psychotic symptoms (Kelleher et al., 2013) • ASD (Duerden et al, 2012) • Early-onset (< 16 years) cannabis use in females (Wilcox et al., 2004) • Conduct, hyperkinetic, and emotional problems in males (Sourander et al., 2009) • Worries about sexuality, anxiety (O’Connor et al., 2009) • Low self-esteem, external attributional style (Martin et al.) • Bullying victimisation (Fisher et al., 2012) • Family conflict, History of NSSI (Brent et al., 2009) • Previous suicide attempt, use of a ‘hard’ method (Hulten et al., 2001) • Rehospitalisation (Czyz et al., 2016) • Childhood abuse (Wan et al., 2015) • History of sexual abuse, family self-harm (O’Connor et al., 2009) • Living in a non-intact family (Sourander et al., 2009) • Low level of education (Brunner et al., 2007)From suicidal thinking to suicide attempts • Presence of psychiatric disorders • Body dissatisfaction • Female gender • Hopelessness • Lower IQ • Exposure to self-harm in both • Higher impulsivity friends and family • Higher intensity seeking • Smoking • Non-cannabis drug use • Lower conscientiousness • A greater number of life events Mars et al. (2019)Risk factors for completed suicide • Male sex • Low socioeconomic status • Restricted educational achievement • Parental separation or divorce • Parental death • Adverse childhood experiences • Parental mental disorder • Family history of suicidal behaviour • Interpersonal difficulties • Mental disorder • Drug and alcohol misuse Hawton et al. (2012) • HopelessnessLong term follow up A&E presentations Death in 1% (50% suicides) Repetition in 27.3% • Age • The method used was usually different to that used for self-harm. • Self-cutting • Previous self-harm • Male gender • Self-cutting • Psychiatric treatment • Prior psychiatric treatment • History of previous self harm Hawton et al. (2012) • Violent versus non-violent self-harm makes you 8 times more likely to die Beckman et al. (2019)Peer-adultnetworkstructureandsuicideattempts in38highschools:implicationsfornetwork- informedsuicideprevention • School networks could provide the relationship network structure that will potentially prevent suicidal behaviour • Lower peer network integration and cohesion in schools had higher rates of suicidal ideation (SI) and suicide attempts (SA) • Suicidal attempts increased with two factors: 1. Student isolation - SA rate on averageisolated from adults led to 20% higher 2. Popularity of student and clustering on network - Higher relative to non suicidal peers Wyman et al. (2019) Riskassessment:future Implicit associations https://implicit.harvard.edu/implicit/user/pimh/ind ex.jsp Machine learning https://www.nature.com/articles/s41398-020- 01100-0Stressors in current pandemic • Anxiety and fear relating to the pandemic (Guessoum et al., 2020) • Isolation, loneliness (Reger, Stanley, and Joiner, 2020) • Pre-existing mental illness (Moutier, 2020) • Access to mental health services (Fegert et al., 2020) • Socio-economic disadvantages (Fegert et al., 2020) • Domestic violence (Bradbury-Jonesand Isham, 2020) • Alcohol consumption (Dumas, Ellis, and Litt, 2020) • Increased exposure to social media (Xiong et al., 2020, Sedgwick et al., 2019) • Bereavement (Clemens et al., 2020)Retrospective Cohort study: Methodology • First and to date the only international study on self-harm in children and adolescents • Electronic patient records • Emergency unit presentations (n=2073) • March–April 2019 & March–April 2020 • Under-18s • Psychiatric emergencies including self-harm (Ougrin et al., 2020, under review)Catchmentareas • 10 countries • 23 hospital A&E • 6.5 million children and adolescents • Mixture of health care systems • Categorised into 14 areas for analysesMain results • No. of emergency psychiatric Total number of emergency psychiatric presentations presentations decreased significantly 1400 • 1,239 in 2019 → 834 in 2020 1200 • IRR = 0.67, 95%CI [0.62, 0.73] 1000 800 • Proportion of self-harm 600 400 presentations increased significantly 200 • 50% in 2019 → 57% in 2020 0 • OR = 1.33, 95%CI [1.07, 1.64] 2019 2020 SH non SHMain results Forest plot of year differences in hospital self-harm presentationsResults • Proportion with history of previous hospital presentation for self-harm • Significantly increasedin 2020 • OR 1.40, 95%CI [1.05, 1.87] • Proportion with history of previous self-harm in community • No significant difference in 2020Results: Clinical characteristics Among those presenting with self-harm, the proportion of… • Severe self-harm* • No significant difference in 2020 *High-lethality method, ICU admission,or Acute ward for >72 hoursResults: Clinical characteristics Among those presenting with self-harm, the proportion of… Proportion presenting with diagnosis of emotional disorder • Emotional disorder diagnosis 70% • Increasedsignificantly in 2020 60% • OR 1.58, 95%CI [1.06 to 2.36] 50% 40% 30% 20% 10% 0% 2019 2020Results: Clinical management Among those presenting with self-harm, the proportion of… Proportion admitted to observation ward 14% • Admission to observation ward 12% • Reduced significantlyin 2020 10% • OR 0.52, 95%CI [0.28 to 0.96] 8% 6% 4% 2% 0% 2019 2020Notable negative results Among those presenting with self-harm, no statistically significant difference was found in 2020 for the proportion… • From deprived areas • From ethnic minorities • Offered follow-up appointments • Subsequently attended the first follow-up appointmentImplications of findings • Comparing with inpatient psychiatric admissions in England… Data retrieved from the National Commissioning Data Repository (NCDR)Evaluate effects of lockdown policies • OxCGRT (Hale et al., 2020) • Daily, standarised, and country-specific measure of lockdown stringency • Nine policy response indicators • School closure • Workplace closure • Public event cancellation • Restrictions of gatherings • Public transport closure • Stay at home requirements • Restrictions on internal movement • International travel controls • Public info campaignsSecondary analyses • Mediation roles of lockdown stringency on… • Reduction in emergency psychiatric presentations • Reduction in self-harm presentations • Increase in proportion of self-harm presentations • Lockdown stringency as a predictor • Characteristics of children and adolescents presented with self-harm during March and April 2020Results: Mediation effects of stringency IRRNI= 0·44 95% CI [0·37, 0·51] Lockdown Stringency No. of emergency Year CDE psychiatric presentations IRR = 1·32 95% CI [1·13, 1·58] Total effect: IRR = 0·58, 95% CI [0·53, 0·65]Results: Mediation effects of stringency IRRNI= 0·51 95% CI [0·41, 0·63] Lockdown Stringency No. of self-harm Year CDE presentations IRR = 1·34 95% CI [1·07, 1·68] Total effect: IRR = 0·68, 95% CI [0·60, 0·80] Results: Contrasting patterns across deprivation levels • When lockdown became more stringent, children from more deprived neighbourhoods became less likely to be presented for self- harm. • However, they were not always less likely to be presented when compared directly with peers from relatively more deprived deciles. (e.g. 3 vs 7 decile) More deprivedSummary • Suicide is the second leading cause of death in young people • Self-Harm is the strongest predictor of death by suicide • DBT-A is the leading treatment for young people with self-harm • Self-Harm presentations are reduced in the initial phases of pandemics • Lockdown stringency mediates the reduction in psychiatric emergency and self-harm presentations in 2020 compared to 2019