A small body of previous research has found that when children are admitted to hospital as a result of serious injuries or poisonings (self-inflicted, perpetrated, or accidental), such trauma-related hospital admissions are associated with future risk of self-harm, suicidality, and violent criminality (Finkelstein et al, 2015; Ballard et al, 2015).

There is a need to better understand the long-term trajectories of these children, from point of hospital admission into adult life, to inform preventive initiatives geared towards reducing the risk of later poor outcomes in these individuals and the healthcare costs associated with this.

Consequently, new research recently published in The Lancet Public Health, led by Dr Roger Webb (2017), has sought to examine the extent of risk of self-harm and violent criminality in adolescence and young adulthood following a trauma-related hospital admission during childhood, within a large national cohort of individuals in Denmark.

Methods 

The study cohort (N = 1,087,672) was drawn from the Civil Registration System in Denmark and consisted of all people born between 1977 and 1997, who had Danish-born parents, and who were still living in Denmark by their 15^th birthdays. Trauma-related hospital admissions for injuries or poisonings occurring between individuals’ births and their 15^th birthdays, and for self-harm occurring between individuals’ 10^th and 15^th birthdays, were identified through data linkage with the National Patient Register and the Psychiatric Central Research Register.

The follow-up period for the cohort was the time between individuals’ 15^th and 35^th birthdays. Hospital admissions as a result of self-harm during this follow-up period were similarly identified through data linkage with the National Patient Register and the Psychiatric Central Research Register. Violent crime convictions were identified through data linkage with the National Crime Register. Parental socioeconomic status (SES), as identified through data linkage with the Integrated Database for Labour Market Research, was also measured as a potential confounding influence on the relationships between childhood trauma-related hospital admission(s) and later self-harm and/or violent criminality.

Results

The authors found that 10% of the cohort had experienced a childhood trauma-related hospital admission (105,753; men: 64,454 [11%]; women: 44,299 [8%]), with ‘accident’ being the most common reason recorded for this and ‘interpersonal violence’ being the rarest reason.

Incidence rate ratios (IRRs) showed that having at least one childhood trauma-related hospital admission was associated with a significant increase in risk for subsequent self-harm and/or violent criminality during the follow-up period (age 15-35 years).

Stronger associations were found when:

• The childhood hospital admission was the result of self-harm or interpersonal violence, as compared to when it followed an accident
• The individual who had experienced the childhood trauma-related hospital admission was female (although similar patterns in general were found for men and women). For example, for women who had been admitted to hospital during childhood following interpersonal violence, risk for subsequent violent criminality among these women increased ten-fold.

While the prevalence of repeat trauma-related childhood hospital admissions was generally low across the cohort, IRRs showed that having three or more childhood hospital admissions (as compared to none) was associated with risk for both adverse outcomes, as was being admitted to hospital for more than one reason. Again, these associations were much stronger for women than men.

In terms of levels of absolute risk, the authors found that this was particularly high for:

• Violent criminality in young adult men who had experienced a childhood hospital admission as a result of interpersonal violence.
• Self-harm in young adult women who had experienced a childhood hospital admission for self-harm or interpersonal violence.

Parental socioeconomic status appeared to have little influence on any associations.

Strengths and limitations

The major strength of this study lies in the fact that through their use of data linkage between multiple national registers in Denmark, the authors attenuated several issues that can plague longitudinal studies of this nature, including sample attrition, participant recall bias, missing data, and low statistical power or precision to measure fairly rare events.

In terms of the main limitations of this study:

• The authors only included hospital admissions for self-harm as their proxy for the self-harm adverse outcome during the follow-up period. This means that their study could have underestimated self-harm rates within the cohort, given that individuals who self-harm often do not present to health services (RCPsych, 2014).
• The authors did not include emergency room and ambulatory care visits in their criteria for a trauma-related childhood hospital admission. Previous research has shown that only approximately half of visits to the accident and emergency department for self-harm actually result in an inpatient hospital admission for young people (Olfson et al, 2005). However, as the authors acknowledge, inclusion of this could also have meant that a relatively large amount of minor (and thus likely less traumatic) childhood accidental injuries would have been included in their sample.
• As the authors describe, an inherent limitation of studies of this nature is the issue of residual confounding, essentially meaning that despite trying to control for as many potential confounding variables as possible, there will always be those not taken account of that could influence the findings. For example, parental mental health issues and bullying by peers are also risk factors for self-harm and suicide in young people (Hawton et al, 2012).
• Unlike in previous research in this area, the authors could not include individuals in their sample whose hospital admission had been classified as being of ‘undetermined cause’, due to unavailability of this data in the registration system. This previous research has shown that these individuals also have increased risk for subsequent suicidality (Connor et al, 2003).

Conclusions and implications

This study has highlighted the risk that a trauma-related childhood hospital admission (defined in terms of admission for self-injuries or poisonings) can confer on two adverse outcomes for young adults: self-harm and violent criminality. In their commentary on this study in The Lancet Public Health, Grant and Lappin highlight:

That there is an association with later self-harm or violence is perhaps not surprising, but the strength of the associations is striking
(Grant and Lappin, 2017).

Ultimately, the authors found that for young adults:

• Men who had experienced a childhood hospital admission as a result of interpersonal violence
  • a fifth will have committed a violent crime by their 25^th birthday and
  • a quarter will have done so by their 35^th birthday;
• Women who have experienced a childhood hospital admission for self-harm or interpersonal violence
  • a fifth will have been re-admitted to hospital for self-harm by their 25^th birthday.

In terms of the implications of the findings of this study, the findings crucially suggest that:

The wellbeing of individuals who experienced hospital admission during childhood for self-harm or interpersonal violence merits especially close attention.

Indeed, the findings arguably have important implications for the clinical care of these individuals, as well as for the preventive initiatives that could be put in place to try and safeguard the wellbeing of such individuals. Specifically, clinical care and preventive initiatives could need to address the:

1. Antecedents of such a hospital admission, particularly when the cause of the admission was interpersonal violence, which could be achieved as the authors suggest through conducting a family-level psychosocial assessment and necessary follow-up; and
2. Possible consequences of such an admission, such as through the delivery of targeted approaches (e.g. counselling) or whole school approaches to the promotion of mental health.

Links

Primary paper

Webb, Roger T et al. (2017) Self-harm and violent criminality among young people who experienced trauma-related hospital admission during childhood: a Danish national cohort study. The Lancet Public Health DOI: http://dx.doi.org/10.1016/S2468-2667(17)30094-4

Other references

Finkelstein Y, Macdonald EM, Hollands S, Hutson JR, Sivilotti MLA, Muhammad MM, Juurlink DN. (2015) Long-term outcomes following self-poisoning in adolescents: A population-based cohort study. The Lancet Psychiatry, 2, 532-539. [Abstract]

Ballard ED, Kalb LG, Vasa RA, Goldstein M, Wilcox HC. (2015) Self-harm, assault, and undetermined intent injuries among paediatric emergency department visits. Pediatric Emergency Care, 31, 813-818.

Royal College of Psychiatrists. (2014) Self harm. Website last accessed 10 Aug 2017. http://www.rcpsych.ac.uk/healthadvice/problemsdisorders/self-harm.aspx

Olfson M, Gameroff MJ, Marcus SC, Greenberg T, Shaffer D. (2005) Emergency treatment of young people following deliberate self-harm. Archives of General Psychiatry, 62, 1122-1128.

Hawton K, Saunders KEA, O’Connor RC. (2012) Self-harm and suicide in adolescents. The Lancet, 379, 2373-2382. [Abstract]

Conner KR, Langley J, Tomaszeski KJ, Conwell Y. (2003) Injury hospitalization and risks for subsequent self-injury and suicide: A national study from New Zealand. American Journal of Public Health, 97, 1128–1131.

Grant S, Lappin J. (2017) Childhood trauma: Psychiatry’s greatest public health challenge? The Lancet Public Health. DOI: 10.1016/S2468-2667(17)30104-4

Photo credits

• Photo by Glen McCallum on Unsplash