The Challenge and Complexities of Physical Abuse

As a practising paediatrician, it is my impression that, in spite of overall increases in the volume of child protection work, we are actually seeing fewer cases of physical abuse, particularly in relation to severe and fatal physical abuse. If that is indeed the case, it is a cause for celebration, as it implies that we are more effectively protecting children from physical abuse, and perhaps that our societies are becoming less tolerant of physical violence to children. Several publications over recent years support this view, but it is not a uniform or consistent finding. Gilbert and colleagues (2012) examined trends in child maltreatment rates in six developed countries, and found a mixed picture, with stable or falling rates of violent child deaths in infants and children in all six countries. In contrast, only Sweden and Manitoba showed decreasing rates of maltreatment-related injury admissions. More detailed analyses of violent deaths in England and Wales showed decreases in both infants and children (Pritchard and Sharples, 2008; Sidebotham et al. 2012). Reporting on cross-sectional population surveys in the USA, Zolotor et al. (2011) showed promising downward trends in the use of physical punishment (though not in the use of objects for hitting a child). The fourth US National Incidence Study found decreases in the rates of physical abuse of 23–29 per cent compared to the previous study a decade earlier (Sedlak et al., 2010). A 2009 study in the UK showed that fewer 18–24-year olds reported physical abuse than in a similar study in 1998 (Cawson et al., 2000; NSPCC, 2011). In contrast, researchers in The Netherlands found no change in the rates of physical abuse between 2005 and 2010 (Euser et al., 2013). ‘Perhaps our societies are becoming less tolerant of physical violence to children’ ‘In contrast, researchers in The Netherlands found no change in the rates of physical abuse’ In understanding prevalence rates for maltreatment, the work of the Centre for Child and Family Studies in Leiden is exemplary (Stoltenborgh et al., 2011, 2013a, 2013b). In this issue of Child Abuse Review, Marije Stoltenborgh and colleagues (2015) from the centre report on a series of meta-analyses that they have undertaken collating data from across the globe on all forms of maltreatment. They found an overall global prevalence of 226/1000 for self-reported physical abuse, and three per 1000 for informant reported. Self-reported prevalence varied by geographical region and between countries. Another paper in this issue looks at the prevalence of maltreatment in Nordic countries, systematically searching the published literature from 1990 (Kloppen et al., 2015). They found self-reported prevalence rates of severe physical violence of between three and nine per cent, and mild physical violence of between 13 and 72 per cent. Of note, studies in Finland and Sweden showed lower rates in more recent studies compared to those in the 1990s. These findings highlight that, even if there have been some improvements, child physical abuse remains a major concern that continues to affect the lives of as many as one in five children around the globe, and presents ongoing dilemmas for professionals in relation to recognition, diagnosis and management. The mismatch between self-reported and informant-reported prevalence rates in Stoltenborgh et al.’s study shows that the vast majority of physical abuse goes unrecognised by professionals. So how do we improve our recognition and response? Faced with a child with an injury, professionals may struggle to conclude whether the injury is unintentional, ‘accidental’; whether it is a result of physical abuse; or whether it may indicate a broader picture of neglect or violence in the home. ‘Child physical abuse remains a major concern that continues to affect the lives of as many as one in five children’ The Welsh systematic review group (http://www.core-info.cardiff.ac.uk) in Cardiff has provided some excellent resources to support clinicians in this task (Kemp et al., 2008, 2014; Maguire and Mann, 2013; Maguire et al., 2009, 2013). In this issue, Ben Cosway and colleagues (2015) from the same group report a case series which will be of help to clinicians presented with an unexplained rib fracture in a young child. They concluded that occult rib fractures have a high probability of abuse (accounting for one-third of their cases). While they concluded that the number, location and distribution of rib fractures are not helpful in distinguishing abuse from other medical or non-intentional causes, other features can be. All the children in their series with a diagnosis of physical abuse had other features of inflicted injury or neglect, and all infants with metabolic bone disease of prematurity had at least one risk factor for metabolic bone disease in addition to their prematurity. They argue for a comprehensive clinical assessment of all infants presenting with rib fractures and provide a helpful template for this assessment. ‘All the children had other features of inflicted injury or neglect’ In another paper from the Welsh group, Zoë Lawson and colleagues (2015) report on the consistency of measuring bruises from photographic images. They found that there was considerable variation both between observers and over time, using both electronic and manual methods of measurement. Their data suggest that 95 per cent of measurements for any particular bruise will lie within a range of 2 cm. However, as the authors point out, the recorded contours and size of a bruise will be affected not only according to observer variation, but also by numerous other factors including the depth of the injury, the force exerted and the time interval from when it was sustained. While for some injuries, particularly those with specific imprints, exact measurements may be extremely important, for most bruises, the ability to measure, and the clinical relevance of accurately measuring, a bruise to within less than 2 cm is questionable. Common sense and caution are perhaps as important here as rigour and detail. ‘Common sense and caution are perhaps as important here as rigour and detail’ In the next paper in this issue, Brennan Wilson and Hilary Smith (2015) present an extremely helpful case review of the histories given by parents of young children with abusive long bone fractures. The authors undertook a textual analysis of medical and legal records in 18 children presenting through the courts. In these 18 cases, the authors identified a total of 41 accounts provided by the carers to explain the injuries. They identified two primary types of account which were found almost without exception. Type one accounts consisted of minor incidents happening spontaneously and without provocation during the normal care of the child. These accounts were considered incompatible with the injury in terms of timing, mechanism or the force required. Type two accounts consisted of confessions or accusations of rough handling of the child. Again, these were not thought to be compatible with the injuries observed. In all cases, the accounts were often vague or uncertain, and frequently multiple accounts were given as the injuries came to light. Drawing on their analysis, Wilson and Smith provide a ‘typical’ presentation of a baby or young child with a long bone fracture: ‘In all cases, the accounts were often vague or uncertain, and frequently multiple accounts were given as the injuries came to light’ ‘The child is typically brought to his general practitioner or to the local accident and emergency department by his principal carers. On arrival, the carers state that they have come because they have noticed signs and symptoms which the health professionals attribute to a major injury…’ (p. 23) It is worth reading the whole account and being alert to the features that they highlight – there is much to learn here. As with much of child protection work, however, some caution should be exercised. While it may be possible to draw similarities between cases, and to highlight typical findings, the very nature of child maltreatment is such that complexity exists. While many cases may fit a classic presentation, others will not, and there can be multiple reasons for the manner in which cases present. In clinical practice, it is not unusual to obtain a changing history. Many factors can influence this, including the person who is taking the history, their experience and the questions they ask; how they record their findings; the person giving the history; the context and location of the interview; fear or anxiety in the parent or carer; and many other contextual factors. Finding our way through this complexity requires an authoritative approach, combining a thorough understanding of the circumstances and context of the case, with an appraisal of the evidence base, the practitioner's own expertise and experience, and the humility to work in partnership with children, their parents or carers, and other professionals (Sidebotham, 2013). ‘In clinical practice it is not unusual to obtain a changing history’ Our final paper in this issue highlights another aspect of complexity in decision-making in physical abuse. Amanda Stephens and Kim Oates (2015) studied the outcomes of a cohort of 68 children with suspected non-accidental head injury in Sydney. They found very high rates of re-notification to child protection services, regardless of placement type. Despite the fact that the majority of children were removed from their families, large numbers of them continued to experience inadequate or inappropriate care. ‘Despite the fact that the majority of children were removed from their families, large numbers of them continued to experience inadequate or inappropriate care’ In the Australian context, this is particularly an issue in relation to the indigenous population where, in spite of appropriate policies promoting placement within their cultural context, a shortage of foster carers can make this difficult to achieve. The authors point out that non-accidental head injury, ‘like all forms of physical abuse, is the result of the complex interaction of a number of child, parental and environmental factors’ and that ‘these complexities present challenges for decision-makers’ (p. 73). They call for policy discussions to reconsider approaches to primary and secondary prevention, taking account of these complexities. Returning to the papers by Stoltenborgh et al. and Kloppen et al., it seems that there is a need for some quite profound policy discussions extending beyond the bounds of non-accidental head injury, or even of physical abuse, and not just in Australia, but across countries around the globe. These discussions need to involve policy makers, practitioners and academics, and,crucially, need to be informed by strong engagement with children and young people and with their carers. Both papers highlight dramatically the paucity of research into emotional abuse and neglect. Of 551 prevalence rates reported in the literature reviewed by Stoltenborgh et al. (2015), only 46 were for emotional abuse, 17 for emotional neglect and 15 for physical neglect. Similarly, in the Nordic review, only one of 24 papers reported on the prevalence of neglect and emotional maltreatment (Kloppen et al., 2015). This is despite the fact that estimates of emotional abuse prevalence are generally of an order of magnitude higher than those of sexual or physical abuse. Given all that we know of the long-term impact of neglect and emotional abuse, it is time that we took this seriously both in research and in policy terms (Brandon et al., 2014; Turney and Taylor, 2014). ‘Both papers highlight dramatically the paucity of research into emotional abuse and neglect’ The papers by Stoltenborgh et al. and Kloppen et al. may give some cause for celebration around decreasing prevalence figures for physical abuse. However, they certainly don't give any reason for complacency – either in relation to physical abuse itself, or in the wider picture of other forms of maltreatment. In all these areas, we work in a complex field where challenges remain for practitioners, academics and policy makers alike.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.