Introduction
The international literature is beginning to focus on child and adolescent mental health presentations to the emergency department (ED). Many authors highlight the resource burden of paediatric emergency admissions resulting from such presentations (Case et al. Reference Case, Case, Olfson, Linakis and Laska2011; Mahajan et al. Reference Mahajan, Alpern, Grupp-Phelan, Chamberlain, Dong and Holubkov2009, Kehoe & McNicholas, Reference Kehoe and McNicholas2018) and make suggestions for improved resource allocation to divert these patients from the ED. Others focus on developing best practice in suicide/self-harm risk assessment or qualitatively evaluate service user and carer experience in the paediatric ED setting. The bulk of research to date tends to focus on adolescents, particularly as the 15–19 age category is a recognised or ‘peak’ risk group for deliberate self-harm (DSH) (Rodway et al. Reference Rodway, Tham, Ibrahim, Turnbull, Windfuhr, Shaw, Kapur and Appleby2016; McMahon et al. Reference McMahon, Reulbach, Corcoran, Keeley, Perry and Arensman2010).
A generalised trend in increased psychiatric presentations in children and adolescents has been noted in multiple jurisdictions in recent years – 47% over 11 years in one American paediatric ED (Mapelli et al. Reference Mapelli, Black and Doan2015), threefold in a Copenhagen paediatric ED over 10 years (Taastrøm et al. Reference Taastrøm, Klahn, Staal, Thomsen and Johansen2014), 6.5% per year in an Australian paediatric ED (Hiscock et al. Reference Hiscock, Neely, Lei and Freed2018) and threefold in a Belgian ED over a 5-year period (Cuypers et al. Reference Cuypers, Danckaerts, Sabbe, Demyttenaere and Bruffaerts2014). A recent Irish study showed a 526% increase in mental health presentations to a paediatric ED over a 10-year period (Fitzgerald et al. Reference Fitzgerald, Foley, McNamara, Barrett, Boylan, Butler, Morgan and Okafor2020).
Regarding children (12 years and under) specifically, less is known. It is posited that children may have different emotional needs compared with adolescents, for example, a higher prevalence of conduct disorder symptoms and different methods of self-harm (Preyde et al. Reference Preyde, Watkins, Csuzdi, Carter, Lazure, White, Penney, Ashbourne, Cameron and Frensch2012). One Irish study highlighted an increase in children presenting over a 6-year period to paediatric EDs with self-harming behaviours (Sarkar et al. Reference Sarkar, Byrne, Power, Fitzpatrick, Anglim, Boylan and Morgan2010) and suggested that children are more likely to report bullying as a precipitant. A 10-year review of the US psychiatric presentations in 6–20-year-olds (Simon & Schoendorf, Reference Simon and Schoendorf2014) demonstrated a continuous year-on-year increase in presentations, from 4.4% of all ED visits in 2001 to 7.2% in 2011. While there was no statistically significant upward trend in presentations for under-12s compared with the adolescents, this study excluded a possibly increasing cohort of children presenting with emotional difficulties or challenging behaviours in the context of autistic spectrum disorder (ASD). Another large Canadian retrospective study noted that under-12s were less likely to return to ED for mental health reasons compared with 13–17-year-olds (Newton et al. Reference Newton, Ali, Johnson, Haines, Rosychuk and Keaschuk2010). An earlier 7-year review across the United States also demonstrated an overall increase in presentations by children and noted this was primarily attributable to patients with behavioural disorders, rather than emergent psychiatric diagnoses or suicide attempts per se (Sills & Bland, Reference Sills and Bland2002).
In light of the limited evidence for this age group, this study aims to review psychiatric presentations of children (aged 0–12 years) to Tallaght University Hospital over a 10-year period.
Methods
Setting
Children’s Hospital Ireland at Tallaght University Hospital (formerly the National Children’s Hospital) provides psychiatric services to children and adolescents aged 15 years and under. Referrals to the CAMHS Liaison Service are made by paediatricians. From 2016, referrals were accepted directly from the ED as well as from the paediatric ward. The service also provides assessments for children and adolescents who have been medically admitted to the paediatric wards or who are attending the paediatric outpatients department. The service provides assessments between 9 a.m. and 5 p.m., Monday–Friday, and at weekends between 9 a.m. and 7 p.m. Children attending the ED after hours and who require urgent psychiatric assessment are admitted to the paediatric ward to facilitate assessment the following day.
Data collection
A retrospective study of case notes was undertaken of all children referred to the CAMHS Liaison Service, who were aged 12 years and under on the date of presentation between 1st January 2009 and 31st December 2018. Individual discharge summaries were reviewed from their clinical records. All children identified had a detailed summary available to access. Variables of interest were recorded on a secure database on a hospital encrypted computer. No identifying information was recorded, ensuring all data collected were anonymised.
Variables
Demographic information regarding the age and sex for each presentation was recorded. Prior diagnoses of an ASD or intellectual disability (ID) were recorded. Information on each presentation was recorded, namely, primary presenting complaint (reason for referral), requirement for admission and (if admission required) duration of admission in days. Repeat presentations were recorded. Due to the acute nature of a CAMHS Liaison Service, the primary presenting complaint was derived from the primary reason for referral contained within the case note. Primary presenting complaint was categorised according to the chief impression and/or formulation described by the assessing mental health professional (Consultant in Child and Adolescent Psychiatry, Registrar in Child and Adolescent Psychiatry or Clinical Nurse Specialist in Child and Adolescent Psychiatry) therein. All assessments were discussed with a Consultant in Child and Adolescent Psychiatry before discharge. The primary presenting complaint was not a diagnosis, as we provide an acute assessment service only. Some children may have had a formal diagnosis made (e.g. by a Community CAMHS team) which was known to us at the time of assessment, but we record only the primary presenting complaint in our data. Suicidal thoughts and/or behaviours were noted and further categorised according to method where self-harm occurred as part of the presentation.
Measures of social stressors were recorded, including if referral to medical social work was completed, and if the children were subject to a care order. Finally, information on discharge planning and recommendations was noted. Total referrals to the CAMHS Liaison Service by year were also noted; however, no further details on children greater than 12 years were recorded.
Data analysis
Data were analysed using SPSSv25. Categorical data were displayed as totals and percentages. Pearson’s correlation was used to analyse whether there was a significant positive relationship (increase) in presentations with increase in years. It was also used to analyse whether there was a significant change in different presentations and methods of self-harm over time. Chi-squared tests were used to compare categorical variables. Where significance on chi-squared tests was noted, adjusted residual values were used to determine the nature of the association between variables, with a value of great or equal to 2 denoting a positive statistical significance (Agresti, Reference Agresti2003).
Results
Three hundred and eighteen presentations of children aged 12 years and under were identified using the above method. However, 15.4% (n = 49) were repeat presentations. Fewer than 5% presented more than twice; however, repeat presentations were not counted after the child reached age 13. It is therefore possible that some children continued to present into their adolescence.
Table 1 depicts all cases categorised by primary presenting complaint. Overall, the most common primary presenting complaint was categorised as disturbance of conduct and/or emotions, that is, most consistent with symptoms of emotional dysregulation, conduct/oppositional defiant disorder type symptoms or challenging behaviours. The most common type of mental disorder symptoms presenting was anxiety symptoms, as expected in this age group, followed by depression symptoms.
Presentations of under-12s to a CAMHS Liaison Service over a 10-year period
ADHD, Attention deficit hyperactivity disorder; ASD, autistic spectrum disorder; ID, intellectual disability.
There was a rise in total referrals of children from 2016 onwards. The proportion of children referred has risen from 10% of total referrals in 2009 to 24% of total referrals in 2018 (Fig. 1).
Number of Presentations to a CAMHS Liaison Service (all ages) 2009–2018.
Table 2 depicts the increase in presentation over the 10-year period studied, categorised by primary presenting complaint. Pearson’s correlation showed a significant positive correlation between the total number of presentations and number of years (r(8) = 0.66, p = 0.02). When categorised according to primary presenting complaint, there was a significant increase in children presenting with a disturbance of conduct and/or emotions over time (r(8) = 0.79, p < 0.001).
Presentations of under-12s to a CAMHS Liaison service over a 10-year period, by primary presenting complaint
The median age of all presentations from January 2009 to December 2018 was 11 years. The mean age was 10.5 years. 55% (n ea175) were male and 45% (n 75143) were female. Five per cent of our studied population had known comorbid diagnosis of ID, whereas 15% of our studied population had a known diagnosis of ASD. Eight per cent had a noted diagnosis of dyslexia and/or dyspraxia.
Primary presenting complaint
Overall, the most common primary presenting complaint category recorded in our assessments was that of disturbance of conduct and/or emotions (55%). We use this umbrella term to encompass emotional dysregulation, conduct/oppositional defiant type symptoms or challenging behaviours within the context of ASD or ID.
The vast majority of the remaining presentations related to the category of mental disorder symptoms, of which anxiety symptoms were the most frequent (16% of presentations), followed by depression symptoms (10%). All primary presenting complaints are described in Table 2.
A further analysis compared presenting symptoms among children with a diagnosis of disability, either ASD or ID compared to those without such a diagnosis at presentation (Table 3). This showed a significant association between children without ASD or ID presenting with anxiety symptoms (X 2 = 4.1, df = 1, p = 0.04). There was also a significant association between those with ASD or ID presenting with disturbance of conduct and/or emotions (X 2 = 3.8, df = 1, p = 0.03). For children with ID, one quarter of presentations related to social needs, including acute care placement difficulties, compared with just 4% of children without disability.
Differences between ASD/ID and non-ASD/ID cases
ADHD, Attention deficit hyperactivity disorder; ASD SDautism spectrum disorder; ID sintellectual disability.
Bold font indicates statistical significance at p < 0.05 level. Note: degrees of freedom are equal to 1 for this analysis.
Suicidal thoughts or behaviours
Over half (53%) of children presented with suicidal thoughts or behaviours, and of this subgroup, over four in ten children engaged in suicidal behaviours. Rates of suicidal thoughts only were similar between children with or without diagnosed disability, whereas suicidal behaviours were more frequent in children without disability.
Overall, 22% of the population engaged in suicidal behaviours. Methods were evenly split between cutting, deliberate ingestion and use of a ligature in attempted strangulation or hanging at approximately 33% each. There were no reported uses of a ligature in attempted strangulation or hanging in this cohort until 2012 (Fig. 2).
Recent emergence of ligature use in cases with suicidal behaviours.
Rates of ligature use increased from 2014 onwards, with 45% of self-harm recorded in 2017 by ligature use. For this subset of children presenting with attempted hanging/strangulation, males outnumbered females by almost four to one. Table 4 demonstrates that ingestion was a more common method of self-harm in females, whereas ligature was a more common method in males. The chi-squared test showed that deliberate ingestion was significantly associated with females (X 2 = 12.73, df = 1, p < 0.05) and ligature use was significantly associated with males (X 2 = 5.54, df = 1, p < 0.05).
Differences between male and female cases
ADHD, Attention deficit hyperactivity disorder.
Bold font indicates statistical significance at p < 0.05 level. Note: degrees of freedom are equal to 1 for this analysis.
The mean age of children engaging in suicidal behaviours was 10.8 years. Of the children who engaged in suicidal behaviours, those with ASD were more likely to present with ligature use as the method of self-harm (n = 5/8) than those without ASD/ID (n = 18/61). No children with ID presented with ligature use.
Only 5% (n = 16) of cases involved alcohol or substance misuse as part of the acute presentation.
A referral to the medical social worker is indicated if child welfare, safety or protection concerns were raised during psychiatric assessment. In our sample, 47% of presentations were referred to the medical social worker. However, 6% (n = 18) of cases concerned a child subject to a formal care order.
The median length of admission was 1 day, reflecting generally prompt assessment and discharge by our service. The vast majority (72%, n = 230) were referred to Community CAMHS for follow-up (Table 5). However, 9% were referred back to their GP for follow-up. Six presentations required transfer to a CAMHS inpatient unit for further treatment, predominantly for ongoing treatment of eating disorders.
Discharge information by primary presenting complaint across all years 2009–2018
Discussion
The CAMHS Liaison Service sees under-16s for acute psychiatric assessment and management with a discharge and follow-up plan. Adolescents (aged 13, 14 and 15 years) form the majority of our referrals and assessments. In a 10-year period (1st January 2009–31st December 2018), the CAMHS Liaison team in Tallaght University Hospital assessed 318 presentations of children aged 12 years and under. The proportion of children referred has risen from 10% of total referrals in 2009 to 24% in 2018.
There was a sharp rise in total referrals of children from 2016 onwards. The overall increase in mental health presentations of children and adolescents is in line with national trends noted in another Irish paediatric ED (Fitzgerald et al. Reference Fitzgerald, Foley, McNamara, Barrett, Boylan, Butler, Morgan and Okafor2020).
It is unclear why more children are presenting to the hospital for acute mental health assessment. There is ongoing coverage of long waiting lists in community CAMHS (O’Halloran, Reference O’Halloran2017) in recent years, which has possibly led to a shift in acute presentations via the ED. Media reporting of death by suicide in an 11- year-old child in January 2016 (O’Brien, Reference O’Brien2017) may have contributed to the increase in presentations from 2016 onwards. While direct referrals were accepted from the ED as well as from paediatric wards and outpatients’ departments from 2016, this does not affect the total referral number but rather would reduce the burden on beds in the paediatric medical ward. The Inbetweeners, a qualitative study of the mental health problems expressed by 12–18-year-olds in Tallaght (McCarthy Quinn & Comiskey, Reference McCarthy Quinn and Comiskey2019), highlights uncertainty among parents and healthcare providers around the first port of call in accessing child and adolescent mental health services as well as long waiting lists, with 177 young people in Tallaght (CHO7) waiting for a CAMHS appointment in June 2019.
A brief review of our data reveals that the median age in children referred to our service is decreasing year on year. Suicidal thoughts were reported in over half of presentations, whereas suicidal behaviours occurred in almost one quarter.
Examining trends in suicidal behaviours is important as children and adolescents who engage in non-suicidal self-injury (NSSI) are 17 times more likely to die by suicide (Morgan et al. Reference Morgan, Webb, Carr, Kontopantelis, Green, Chew-Graham, Kapur and Ashcroft2017). Suicide is the second most common cause of death among children and adolescents worldwide after road traffic accidents (Patton et al. Reference Patton, Coffey, Sawyer, Viner, Haller, Bose, Vos, Ferguson and Mathers2009). While it is generally recognised that DSH/NSSI does not always coexist with suicidal intent, particularly in the adolescent age group (Brunner et al. Reference Brunner, Kaess, Parzer, Fischer, Carli and Hoven2014), a history of DSH remains the strongest risk factor for completed suicide (Hawton & Harriss, Reference Hawton and Harriss2007; Rodway et al. Reference Rodway, Tham, Ibrahim, Turnbull, Windfuhr, Shaw, Kapur and Appleby2016). One study estimates that suicide rates in under-15s is extremely rare in Ireland (1.6/100 000) (Malone et al. Reference Malone, Quinlivani, McGuinness, McNicholas and Kelleher2012); however, it is evident that self-harm in children and adolescents is on the rise both nationally (Griffin et al. Reference Griffin, McMahon, McNicholas, Corcoran, Perry and Arensman2018) and in other jurisdictions (Pittsenbarger & Mannix, Reference Pittsenbarger and Mannix2014,;Mapelli et al. Reference Mapelli, Black and Doan2015; Taastrøm et al. Reference Taastrøm, Klahn, Staal, Thomsen and Johansen2014; Hiscock et al. Reference Hiscock, Neely, Lei and Freed2018; Simon & Schoendorf, Reference Simon and Schoendorf2014).
Methods vary in frequency worldwide with peaks and falls depending on fads such as the well-publicised asphyxiation ‘game’ in the early 2000s (Centers for Disease Control and Prevention, 2007). One study has suggested that younger adolescents (under-15s) are 3 times more likely to die by hanging versus older adolescents (aged 15–19 years) (Grøholt et al. Reference Grøholt, Ekeberg, Wichstrøm and Haldorsen1998). We discovered a worrying increase in the use of ligatures to enact self-harm in the form of attempted hanging or strangulation children aged 12 years and under (mean age 10.8 years). This occurs in males almost 4 times more frequently than females and is seen in presentations of conduct disorder/oppositional defiance disorder type symptoms or emotional dysregulation, suggesting impulsive rather than planned self-harm. Impulsivity is important as children are probably less likely to contemplate a hopeless future and engage in suicidal planning and execution due to a reduced capacity for abstract thinking, compared with adolescents.
In our cohort, ligature use was the most common enacted method of self-harm among children with ASD. It is possible that the cognitive styles and associated anxiety and deficits in emotional regulation may contribute to the risk in this population group. Ligature use as a method of self-harm is categorised as ‘violent’ and carries an elevated risk of suicide compared to poisoning (Bergen et al. Reference Bergen, Hawton and Waters2012; Runeson et al. Reference Runeson, Tidemalm, Dahlin, Lichtenstein and Langstrom2010). We, and others (Briere & Gil, Reference Briere and Gil1998; Lowenstein, Reference Lowenstein2005), have shown that children may have different psychiatric treatment needs compared with adolescents. One such study (Huefner & Vollmer Reference Huefner and Vollmer2014) undertaken in a residential inpatient unit showed that children were more behaviourally and emotionally troubled than adolescents. More research is needed into the precipitating factors behind acute psychiatric presentations in children versus adolescents. Sarkar & colleagues (Reference Sarkar, Byrne, Power, Fitzpatrick, Anglim, Boylan and Morgan2010) noted that children were 5 times more likely than adolescents to report bullying as a factor in presenting with suicidal phenomena. Today, the increasing prevalence in cyber-bullying is another possible precipitating factor in presentations in this demographic; many of whom we note, anecdotally, have access to social media or group messaging apps.
Some interesting trends were noted in the subset of our population who had ASD and/or ID. In over two-thirds of the ASD population, the primary presenting complaint was disturbance of conduct and/or emotions (68% v. 52% in the non-ASD/ID population). This raises concerns about possible inappropriate presentations to EDs, where acute paediatric or psychiatric liaison assessment services may not represent the most appropriate assessment or treatment service to meet the needs of these children. However, with approximately half of children with ASD presenting with suicidal thoughts and behaviours, this suggests significant psychological stressors in this population, and suggests that service providers may need to review available psychological support networks for children with ASD, to allow targeted appropriate psychological interventions. Similarly, only 14% of children with ID presented with primary mental health symptoms, whereas over a quarter of presentations of children with ID related to social needs including placement breakdown. Further study of the needs of both children with disability presenting to the ED is warranted, to guide service planning and provision for this population into the future.
Limitations
Data were incomplete in four presentations (1.2% of our dataset). Primary presenting complaint could only be approximated from the assessor’s impression of an acute presentation as per the case note on file and did not equate to a formal diagnosis. Furthermore, assessors changed multiple times over the 10-year period studied due to staff turnover. This may affect inter-rater reliability of primary presenting complaints identified. Given the retrospective nature of the study, we are unable to comment on whether, for example, the referrals were all appropriately directed towards mental health or whether the children were later diagnosed with a mental disorder in a follow-up setting. Finally, while we have endeavoured to classify all those children presenting with thoughts of self-harm into ‘suicidal thoughts’ or ‘suicidal behaviours’, we must acknowledge that self-harm covers a wide spectrum of thoughts and behaviours that may not always be so simply categorised.
However, the trends we have identified, specific to children aged 12 years and under, may have important implications for aiding evidence-based allocation of resources and identify gaps in existing CAMHS services that may be leading to increased paediatric ED mental health presentations, particularly as first point of contact. The issue of resource allocation and diversion from the paediatric ED as a point of first contact with CAMHS is topical in many jurisdictions where psychiatric presentations are also on the rise. This may provoke similar discussion here, as recruitment difficulties and growing waiting lists in CAMHS are frequently raised in the national media. Finally, the amalgamation of the current three Children’s University Hospitals in Dublin to form the new National Children’s Hospital will provide an opportunity for more research comparing acute psychiatric presentations in children versus adolescents.
Financial Support
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Conflict of interests
Dr EM has no conflicts of interest to disclose. Dr KG has no conflicts of interest to disclose. CM has no conflicts of interests to disclose. Dr PB has no conflicts of interests to disclose.
Ethical Standards
Ethical approval was granted by the Ethics Committee at Tallaght University Hospital. All data were anonymised and maintained on encrypted devices. There were no active participants in our study as all data were retrospective and descriptive only.
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committee on human experimentation with the Helsinki Declaration of 1975, as revised in 2008.
