This study assessed the quantity of documentation of maltreatment risk factors in injury-related paediatric hospitalisations for two samples of patients, those with a discharge diagnosis indicating definitive or possible child maltreatment and those with a discharge diagnosis of an unintentional injury. The assignment of a maltreatment code relies on documentation in the medical record to indicate that maltreatment is a possibility, and hence it is not surprising that 93% of cases with a maltreatment code had some documented risk factors in the medical record. For the maltreatment code sample, the most commonly documented risk factor was a history of abuse (45%), and other research has found a similar proportion of documentation of abuse history for maltreatment coded cases (41%) [11
]. Being known to the child protection department and the presence of mental health issues in the child or family were documented almost as frequently as the presence of a history of abuse. Parental mental illness, particularly maternal illness has been associated with a higher risk of child maltreatment in previous research [18
Examining the extent of documentation of risk factors in relation to their child protection system status offered an external validation of likely child maltreatment (given that contact with the child protection system is generally due to concerns about the risk of maltreatment). Over 93% of the maltreatment code sample were able to be linked with a record on the child protection system for either previous, current or subsequent maltreatment concerns, though 16% of those that linked to child protection had no documented risk factors in the medical record. Given that these cases were coded as maltreatment related and the broad scope of factors included in the risk factor review this absence of documented risk factors may reflect poor documentation practises.
In contrast, only 11% of cases in the unintentional injury code sample had a risk factor documented in the medical record, despite 32% of the unintentional injury code sample linking with a record on the child protection system. Only 20% of those that linked had one or more risk factors documented in the medical record. Within the unintentional injury code sample, males under 1
year of age were the most likely group to have documentation of risk factors with 44% of cases having some risk factor documented and the most commonly documented risk factor was previous injury-related hospital admission. Children under 12
months are the most vulnerable age group for child maltreatment and epidemiological research has found they frequently comprise the largest proportion of children with abusive injury [19
]. The second most common group to have documentation of risk factors was females aged 15-17
years of age. The most common risk factors documented in this group were alcohol or drug abuse, mental health issues or relationship instabilities. Research has demonstrated that adolescent girls who have been maltreated are at increased risk of illicit drug or alcohol abuse, self harm, including suicide and mental health issues, including depression [20
]. While many of these presentations may have been due to a legitimate unintentional cause of injury, the lack of documentation indicating the consideration of risk factors associated with maltreatment raises concerns that injuries due to maltreatment are not being identified. Given that a higher proportion of children with an unintentional injury code with risk factors documented than without risk factors documented linked to a child protection record in the 12
months post discharge, risk factor documentation in medical records may be used as an early warning of potential harm for children otherwise treated for injuries deemed to be unintentional.
Providing a definitive diagnosis for injury and whether or not it is related to maltreatment can be difficult for clinicians, particularly where parents cannot or will not provide information about the injury event or the information they provide differs from what occurred to cause the injury. The clinician is required to assess the injury and determine whether or not the injury occurred due to a lapse in supervision that may or may not be associated with supervisory neglect, or if a parent or caregiver has deliberately inflicted the injury. Research has shown that parents may provide false information about the reason for the child’s presentation to hospital [21
] and relying completely on that information could result in missing some children who are being maltreated. This serves to reinforce the importance of complete documentation in the medical record. Where the documentation is complete and describes all elements considered in arriving at a diagnosis of maltreatment, or in ruling it out, previous admissions can be referred to and assist in identifying patterns of concerning behaviour or circumstances that can assist in decision making, diagnoses and, where necessary, reporting of maltreatment. Furthermore, mandatory reporting legislation in this state require the reporting of maltreatment based on a reasonable suspicion that harm is occurring, has occurred or may occur in the future [22
] – therefore the documentation required to demonstrate that maltreatment was considered and, where suspected ruled out, is important should the record ever be required in court for legal purposes.
This study does have some limitations. Documentation of risk factors is not a validated measure of existence of these risk factors, but merely an approximation of risk. With limited guidance for clinical staff regarding what factors require documentation in circumstances of maltreatment, documented risk factors are a proxy measure at best. However, it is still important to examine the extent of documentation and the concordance of documentation with child protection records to gauge the likely completeness of documentation practises. While this study has shown some consistent estimates (such as 93% of cases with maltreatment codes having documented risk factors, and a similar proportion of these cases being known to child protection), there were some divergences in the estimates (such as only 11% of the unintentional injury cases having documented risk factors despite almost one-third of this group having a child protection record). Secondly a lack of documentation may not indicate a failure to consider the presence of maltreatment related risk factors, the clinician may simply not have documented this in the medical record. However, this is, of itself, problematic given the importance of the medical record for communication between health staff and its potential use as a source of evidence for prosecution of perpetrators of maltreatment or information for future admissions that may be suspicious for maltreatment.
A potential limitation of this study could be the use of ICD codes to identify maltreatment in the administrative data sets that formed the basis of the study sample. Some research has shown that ICD is likely to under-identify maltreatment if only the definitive codes for abuse are used [23
]. In this research a broad range of ICD codes that included those identified as associated with possible abuse and a sample not associated with abuse were included so the under-representation is likely to be less of an issue than if only definitive codes were used. The research is still likely to under-represent neglect and other forms of maltreatment that are not associated with physical injury in children.
Ethics committees denied researchers the right to collect and analyse data based on the Indigenous status of the child. In Australia, Indigenous children are over-represented in all child protection statistics from notifications through substantiations and out of home care [19
]. This is likely to be important in considering the risk factors associated with maltreatment in this study as well however, the focus of these analyses were the documentation in the medical record. While it is unlikely that documentation would have varied by Indigenous status no analyses were possible to see if there was a variation in documentation according to the Indigenous status of the child.