Child maltreatment (MI) is a major public health and social welfare problem in Australia1 and around the world.2,3 Large numbers of children are exposed, with estimates of 25% in the United States,4 20% in Australia,5,6 and a larger proportion (up to 50%) of the world’s population.3 The consequences of MI are extensive.2,7
MI has profound effects on physical and mental health during childhood and across the life course.7,8 It also impacts economic and social outcomes, such as school participation and achievement9, welfare dependency, addiction, behavioral risky sexual activity and participation in acts of violence.10
The described physiological effects of childhood abuse and neglect on brain development, stress, and inflammatory responses support these observations as causal.11,12
Despite the large body of literature on the impacts of child abuse and neglect, the effect on mortality has received limited attention, especially in adolescence and young adulthood.
The seminal Adverse Childhood Experiences (AEI) study13 showed that adults who self-reported >6 AEs (MI and household dysfunction) died, on average, 20 years earlier than those who did not report any AEs.14 However , only 15% of study participants were under 40 years of age at the time of enrollment.
In a UK study of people born in 1958, researchers reported higher adjusted hazard ratios (aRRs) for death before age 50 in people with >2 AIS compared to those with none,15 and in a Swedish study of people born in 1953, researchers reported an increased risk of death before age 65 in people with some involvement with child protective services (CPS).16 None of the studies reported the specific impact of mortality on early adulthood.
Late adolescence and early adulthood are crucial developmental stages, in which the effects of mental illness, suicidality, and substance use lead to sharp increases in mortality rates.17,18 However, it is known little about whether MI contributes to these premature deaths. With the current study, we aimed to address this evidence gap by examining all-cause and cause-specific mortality from MI exposure in people aged 16 to 33 years, based on the SPIs intervention.
Studies determine MI exposure through (1) surveys of survivors (years or decades after the events), parents, educators, or other service providers or (2)
child protection agency administrative records or death records. All methods suffer from possible sources of error. Survey-based methods are subject to failure to recover (especially from early life events), the distorting effect of community norms, social acceptability bias, survivor bias, and variable thresholds for MI.
The participation of SPIs as an indicator of MI avoids these issues but may also imply imperfect verification of the case. However, in jurisdictions with extensive mandatory reporting of suspected MI, it is unlikely that serious maltreatment will go unreported. The administrative categories of SPIs, designed to ensure a response provided by child protection authorities, can be used as indicators of exposure levels to MI.
Administrative SPI data are really the only MI determination option for exploring the association between MI and death in early adulthood at the population level, given the rarity of the outcome.
| Methods |
> Study design and participants
A retrospective cohort study design was proposed using linked administrative data. The study cohort included all children born in South Australia (SA), Australia, between 1 January 1986 and 31 May 2003, with information obtained from the SA birth registry and collection of perinatal statistics, who survived until the age of 16 (according to the death record), with a total of 331,254 people.
Information was linked across multiple administrative data sets by SA-NT DataLink19 using deterministic and probabilistic algorithms, with extensive administrative review, based on > 50 administrative data sets. A unique identifier was assigned to all cohort members and approved data items were extracted.
The study was part of the Impacts of Child Abuse and Neglect (IANI) research program,6 established to estimate the health and social consequences of MI in AS. Ethical approval for the study was obtained from the Human Health Research Ethics Committee of AS and the Human Research Ethics Committee of the University of AS.
> Study data
• Characteristics of the child and family
The birth registry provided data on each child’s date of birth, sex, maternal age (< 21 or ≥ 21 years), and residential location (at birth), which was mapped to the Index of Relative Socioeconomic Disadvantage (IDSR)20 and assigned to quintiles of disadvantage using Australia-wide cut-points.
The collection of perinatal AS statistics defined maternal marital status (married/common-law, or not married), smoking status (smoker, non-smoker, or unknown), employment status (employed or not employed), all attributes associated with poor outcomes. in children,21 and for the infant, an indicator of serious perinatal and/or congenital problems (still hospitalized 28 days after birth or discharged within 28 days of birth).
• Child protection data
The Department of Child Protection (DPI) data files contained complete records of child protection contacts in AS, including notifications (child protection reports), investigations and grounds from January 1, 1986 to June 30, 2017 and out-of-home care (CFH) placements for all children entered into the system from January 1, 1990 to June 30, 2017, and for children in care as of January 1, 1990, with entry into the system from 1 January 1986.
A child protection notification occurs when the notifier believes that the child is at risk of, or has experienced, serious harm or high-level chronic neglect.22 In AS, most professionals with potential contact with the child (such as health personnel, social workers, educators, child care workers, police, and legal professionals) are required reporters, and are sanctioned if they fail to report.
Child protection notifications are also taken from the public. A child protection investigation is instigated when a notification reaches a designated threshold of concern and is within the remit of the DPI (e.g. family); The outcome options are supported, unsupported, or no finding possible.
Children are placed in CFH when the DPI determines that leaving the child with their family poses an unacceptable risk of serious (further) harm.22 Periods in CFH range from one night to 18 years. For this study, children were classified by age of first entry into child care: before or after age 3, a central axis of development.
The preschool years (3 to 4 years of age) represent a period of dramatic growth in physical, emotional, and cognitive development, with increased mastery of motor skills, impulse control, socialization skills, and language.23,24 The Rapid brain development and relational patterning occur in the first years of life. This cut-off point has been used previously in exploring outcomes associated with MI.9
People in contact with SPIs are heterogeneous in terms of exposure to MI. The DPI adopted a hierarchical categorization that indicated increasing levels of concern to guide its response.22 Individuals were categorized into groups based on the DPI descriptors, plus a subcategorization of children who entered CFH, resulting in 7 mutually exposed groups. exclusive plus no exposure.
Individuals with only notifications were assigned to 1 of 3 groups reflecting the assessment of risk of harm; The people who had been the subject of investigation were assigned to 2 groups depending on whether the alleged abuse had been corroborated; and people who had ever entered CFH were divided into 2 groups depending on the age of first entry.
Death data were obtained from the AS death registry considering all deaths registered in AS from January 1, 1990 to May 31, 2019, noting month and year, ≥ 1 (up to 9) death code(s). 4-digit cause of death (according to the International Classification of Diseases, Tenth Revision [ICD-10]),25 and/or descriptive text fields. Where only a descriptive text field was recorded, the principal investigator and research assistant independently assigned cause-of-death codes, resolving any discrepancies with a third investigator.
A hierarchical classification system was adopted, informed by all recorded cause of death codes. The first assignment was for intentional self-harm (suicide) (ICD-10 codes X60-X84 and Y20); the second was for poisoning, alcohol, drugs, other substances, and mental illness (ICD-10 codes F00-F99, R45, T36-T65, X40-X45, Y10-Y19, and Y48 [but not X60-X84 and Y20]); the third was for motor vehicle crashes and other accidental injuries (ICD-10 codes C01-V99 [but not codes for suicide, substances, or mental illness]); and, finally, the fourth was for natural causes (ICD-10 chapters I to IV, VI to XVIII, XXI and XXII but not the codes of chapters V, IXX or XX).
• Analysis of data
Cumulative deaths per 1000 people ≥ 16 years of age were calculated for each SPI participation category as 1 lower Kaplan-Meier estimator, adjusting the denominator continuously to control through May 31, 2019.
Multivariable Cox proportional hazards regression models were used to estimate aRRs for all-cause mortality. The proportional hazards assumption was evaluated using global proportional hazards tests for univariate and multivariable models, log-log plots, and the scaled Schoenfeld residual plot.
Covariates were included in the model to adjust for potential confounders, assessed at birth, and for any possible impact of MI. Child characteristics included sex, year of birth (1986-1991, 1992-1997, or 1998-2003), and whether the child was still hospitalized or discharged 28 days after delivery. This last variable was chosen as a summary measure of severe neonatal vulnerability, avoiding the need to potentially include highly correlated variables (such as prematurity and low birth weight).
Family measures included maternal age at the time of the child’s birth, quintile of socioeconomic status by area, maternal marital status, maternal smoking, and employment.
Incidence rates of death by cause of death categories were calculated for persons with any contact with SPIs and without contact with SPIs by dividing the total deaths within each cause of death category by total person-years (16 years to May 30, 2019). Poisson regression was used for death counts, compensating for the total time of people at risk, to estimate incidence rate ratios (IRRs). All analyzes were performed using Stata 16.0 (Stata Corp, College Station, TX).
| Results |
> Cohort characteristics
There were 331,254 children in the study cohort, of whom 20% (n = 66,278) had some record of contact with RLS by the age of 16 years, and only 2% (n = 6728) had ever been in CFH. Of those admitted to CFH, 42% (n = 2805) had their first stay in a foster home before age 3 (increasing from 30% to 56% over the study period).
All measures of disadvantage were more common in people in contact with RLS compared to those without contact, with higher percentages of experiencing disadvantage in groups suggestive of exposure to more severe MI. For example, 10% of
children without contact with SPIs had mothers who were not married or in a de facto relationship at the time of their birth, but 20% to 22% of children with a notifier-only concern (PEN) or other notification, 32% to 35 % of children who had been the subject of an investigation (substantiated or unfounded), and 43% to 51% of children who had been admitted to CFH had single or unmarried mothers at the time of their birth.
> Cumulative mortality
Unadjusted mortality rates (per 1000 people) were calculated and graphed from age 16 years and older, according to SPI categories. The highest mortality was experienced by those who had been placed in CFH and whose first admission to care occurred after their third birthday. By 33 years of age, mortality in this group was 30.9 per 1000 compared with 5.1 per 1000 in those without contact with RLS.
Young people who had been the subject of a substantiated child protection notification confirming exposure to severe MI also had considerably higher mortality across the age range than the group without any contact with CLS, reaching 13.7 per 1000 at the age of 33 years. Individuals who had entered CFH before their third birthday experienced lower mortality than those who had entered care after age 3 years.
> Association between RLS participation and mortality
Adjusted and unadjusted risk ratios (RRs) for the risk of death by RLS category were evaluated. In the unadjusted analysis, individuals with any SPI involvement other than PEN had significantly higher RRs than children without SPI involvement (RR = 1.55-5.77). After controlling for sociodemographic characteristics and birth outcomes, aRRs remained significantly elevated in all but one category.
Among those who had entered CFH for the first time after 3 years, the aRR was the highest (4.67; 95% confidence interval [CI] = 3.52–6.20), while for those For people who had entered CFH for the first time before 3 years of age, the RRa was 1.75 (95% CI = 0.98-3.14) compared to the group without contact with the RLS.
For those who experienced a child protection notification (MPI), whether substantiated or unsubstantiated, the RRa was more than double that of the group without CPS contact. None of the indicators of maternal and/or household socioeconomic status at the time of birth of the cohort members were significant in the multivariable model. Being male (compared to female) and remaining in the hospital at 28 days of age conferred more than double the risk of death.
> Cause of death
Causes of death were assessed based on history of contact with SPIs, describing the number of deaths, cause-specific death rates per 100,000 person-years, and RTIs (contact with SPIs versus no contact).
It was found that people in contact with SPIs (indicating MI exposure) were more likely to die in circumstances involving poisoning, alcohol, drugs, other substances, or mental illness (incidence rate ratio [IRR] = 4.82 [ 95% CI = 3.31–7.01]) or suicide (RTI = 2.82 [95% CI = 2.15–3.68]), but also from natural causes (RTI = 1.99 [95% CI] % = 1.52-2.57]), compared to children without contact with SPIs. The largest absolute contributor to excess risk was suicide, at 11.46 per 100,000 person-years (17.77–6.31).
| Discussion |
This is the first study to estimate the impact of MI on all-cause and cause-specific mortality from mid-adolescence to early adulthood using population-level data.
Contact with RLS as an indicator of MI history was found to be strongly associated with an increased risk of death between 16 and 33 years of age. A substantial excess risk of premature death was observed for every SPI contact category except for PEN (a category that does not meet thresholds for concern for MI), after adjusting for a range of potential confounders.
The strength of the association was greater in categories indicating more severe exposure to maltreatment, supporting the relationship as causal, when taken together with well-described mechanisms that relate exposure to MI with altered emotional and behavioral responses, a high sense of shame and low impulse control,11,12,26,27 risk factors for substance use and suicide.
The multivariable model identified an excess risk of death with exposure to MI, which ranged from an aRR of 1.71 (95% CI = 1.18-2.48) to an aRR of 4.67 (95% CI = 3.52–6.20), compared to people without contact with RLS. Looking at heterogeneity within each group (such as exposure to other adversities or protective factors), the strength of the observed associations is notable.
These findings suggest that people with IM exposure have a substantial excess risk of death during late adolescence and early adulthood, possibly greater than the risk differentials at older ages (in a Swedish study16 researchers reported an aRR of 1. 76–2.91 for people in contact with SPIs compared to those without contact, for death before age 65).
This finding is consistent with a view that youth and young adulthood is a period of special vulnerability for people with a history of MI.
For people who had entered CFH for the first time before age 3, effect estimates were smaller (aRR = 1.75) than for people with suspected maltreatment or with confirmed MI who did not enter CFH (aRR = 2.09-2.62) and lower than for people who had entered care after their third birthday (aRR = 4.67).
One possible explanation is that for children who first enter care during infancy, MI exposure is reduced during the early years of rapid brain development and relational pattern formation, limiting potential harms (which is the purpose of out-of-care care). home).
Consistent with this, in the present cohort, people who had entered CFH before age 3 had, on average, 40% more total time in care than children who had entered CFH for the first time after age 3. years. But as noted below, further exploration of the CFH pattern should be warranted.
Distinguishing the impact of MI exposure from SPI involvement presents an ongoing methodological challenge28 and is a crucial area for future studies. It would be valuable, in a future study, to further explore the role of MI exposure (type, age, and relationship to perpetrator), the timing and extent of contact with the child protection system, and for children who enter to CFH, ages of entry and exit from care, total time in care, changes in placement, type of care, family reconnection, and how these characteristics interact by child sex, socioeconomic status, and birth cohort to influence the risk of death.
Regarding causes of death, the largest absolute differential in mortality rates (people with any contact with RLS versus those without any) was related to deaths involving poisoning, alcohol, other substances, mental illness, and suicide. Causes related to mental health problems and/or substance use represented an average of 20.9 deaths per 100,000 person-years, compared to natural causes that contributed 8.2 deaths per 100,000 person-years. This is consistent with extensive evidence linking MI with mental illness, suicide, and substance use.10-12
In the adjusted analysis, being male carried an excess risk of death in youth and young adulthood (aRR = 2.17), underscoring an imperative to do more to support boys and young men, especially those who are victims from my.
In a large American study of adolescents, researchers found that the relative risk of suicide attempts in boys with a history of MI (compared with those without any) was significantly higher than in girls (e.g., for family sexual abuse [aRR = 15.04 for men but aRR = 4.34 for women, with a history of MI versus those without a history]).10
This study has several strengths:
1. First, it is based on data from an entire population, with all births in AS between 1986 and 2003, totaling 331,254 people. This is particularly critical for a mortality study, in which sampling (i.e., a survey in early adulthood) will inevitably exclude the most vulnerable: those who have died, become homeless, been incarcerated, or those with serious addictions.
2. Second, the use of validated, linked, administrative data on MI history, mortality status, and socioeconomic covariates ensures high-quality data.
3. Third, the covariates were collected at the time of the child’s birth, before MI exposure, and as such, are not compromised by their possible location on the causal pathway.
4. Fourth, MI is a complex construct and the use of different SPI participation categories, postulated to indicate the severity of MI exposure and risk of harm, was much more informative.
This study is timely and provides new evidence in an under-researched area, covering deaths recorded up to May 31, 2019. In the context of the escalating costs of child protection globally, due to the administration of CPS and the addressing the consequences of MI,29 (in AS, the costs of SPIs increased from $174 million in 2008-2009 to $504 million in 2017-2018),30 this work generates relevant evidence to guide current policies and shows that A change in practices is crucial.
The study has some limitations. SPI involvement is a strong indicator of probable MI, and yet sometimes maltreatment is not reported to SPIs, and some children who are reported and investigated do not suffer maltreatment, with potential for
misunderstandings, social or racial prejudices. But with occupations that have contact with children, designated mandatory reporters, and protocolized decision rules in use, it was considered that this source of error would be low. SPI participation data were limited to contact with the AS child protection system, and deaths recorded outside of AS were not captured, which represented a small but inevitable source of bias.
The main findings that MI is associated with a substantial excess risk of death during late adolescence and early adulthood and that the more severe the exposure to MI the greater the excess risk, are broadly generalizable.
They are underpinned by strong theory and are consistent with published studies reporting large effects of MI on mental health.8 At the same time, the excess risk of death associated with specific RLS categories could vary across jurisdictions given differences in thresholds. for notification, investigation, justification and suspension of child protection, type of CFH options and policy settings, including access to adoption and early intervention services. Replication of this study in other jurisdictions would be valuable.
| Political implications |
Suicide prevention is a high priority in Australia and internationally. Suicide represents one of the leading causes of premature death and years of potential life lost.31,32 This study highlights the importance of incorporating MI into suicide prevention policy frameworks.
That would add weight to the evidence base indicating that MI is toxic to developing brains and to the creation of an intact sense of self,11,12,33 leading to negative consequences such as increased risk of suicide. A consistently elevated stress response impacts allostatic load and metabolic health. These pathways have implications for mental health and physical health across the lifespan.
The excess risk of suicide and substance-related deaths is considerable, and the resulting death is incontrovertible and potentially preventable. An appropriate response, involving greater support for children and families at risk, is urgently required across clinical and protection services.
Suicide prevention strategies must begin early in life. For pediatricians, primary care physicians, psychiatrists, and other physicians who work with children and adolescents, the need to be alert to the family context of abuse when emotional and behavioral problems are observed is reinforced.12
In other research by the main author34 it was observed that mental health services in AS
are grossly inadequate to support the mental health of infants, children, adolescents and their families. In this mortality study, the tragic consequences of this gap are highlighted. There are evidence-based interventions to improve parent-child relationships,26,35,36 but vulnerable families often do not have access to such programs in Australia or globally.37
The lack of indicators of family socioeconomic status was associated with mortality in the multivariable model (findings consistent with the study by Jackisch et al.16) and, as hypothesized, the observed socioeconomic gradient in health is mainly driven by history of MI.
In studies seeking to describe influences on health disparities, the inclusion of MI as a covariate should be standard.
| Conclusions |
In this study, strong evidence was provided for a substantial excess risk of death in young adults related to MI exposure. This finding reinforces the evidence of the serious consequences of childhood trauma throughout the life course.13
There is no reason to expect that these results do not have broad applicability, given that these findings are a reflection of the underlying physiological pathways between MI and brain development (affecting thinking, behaviors, and emotional responses) and the effect of MI in relational patterns and in the sense of self.
For children who first enter CFH at young ages, SPI participation may reduce risk. AS ECD policies and practices have changed over the time period of this data collection, with a shift toward identifying children at high risk early in life.22 At the same time, there are strong community pressures and considerations ethics that ensure that excluding children from your home is used as a last resort. More evidence needs to be gathered to better understand when child exclusion is beneficial.
The imperative to protect children must extend beyond childhood. The majority of MI-exposed children who come to child protection agencies are not placed in CFH, but outcomes for children who are placed in care, as well as those who are not, suggest that not enough is being done to improve damage or prevent further abuse. Children with suspected MI have a severely increased risk of death in youth and/or early adulthood. And yet, the balance of child protection funds is often centered on CFHs.30
Changes are desired in the service response, including more effective engagement and upskilling of clinicians to provide a coordinated and cross-sector response to childhood trauma.
For pediatricians, who are increasingly seeing children with behavioral and developmental problems,39 this also means being alert to the possibility of toxic stress evidenced by challenging behaviors,12 and recognizing the serious potential consequences if relational trauma is not addressed. As a society, we simply have to do our best to protect children not only from current harms, but also from the extreme consequences of MI over the life course.
