Population-based studies in rich nations estimate that between 4 and 16 % of children under 18 years are abused or neglected each year (Gilbert et al. 2009). While maltreatment rates are mostly stable (Sidebotham et al. 2011a, b) the levels of poverty and deprivation are increasing. In England, for example, the numbers of children living in vulnerable families are expected to more than double between the years 2010 and 2015 (Ofsted 2014). Some families are more vulnerable than others to poverty and this includes families with disabled children. Four out of ten children with disabilities in England live in poverty in comparison with three out of ten children in the general population (Office of the Children’s Commissioner 2013).

More than two decades ago there were arguments that the child protection system which had been designed to remedy individual and psychosocial disorders was instead being used to deal with wider structural problems associated with poverty (Besharov and Laumann 1997; Bebbington and Miles 1989). Besharov argued that intervention in such cases is at best ineffectual and at worst harmful and that in the absence of specialised services, society (in particular the US) ‘would do better if it did nothing in poverty related cases, rather than the wrong – and often harmful- something’ (Besharov and Laumann 1997:5). To improve interventions for maltreated children in poor families, Drake and Pandey argued that it was important to understand the dynamics of maltreatment among those living in poverty (Drake and Pandey 1996).

The recent global recession and widespread austerity measures in the UK to counteract the economic crisis have put the link between poverty and maltreatment into even sharper relief and offered difficult moral choices. For some families, the child protection system represents a starting link in the chain of possible removal of children from their parents. In England, there are renewed claims that the child protection system places a disproportionate burden on those who are already disadvantaged and marginalised (Bywaters 2013). This was demonstrated in the analysis of the socio-economic background of children in the child protection and out of home care system undertaken so much earlier by Bebbington and Miles (1989). However this analysis has been allowed to lapse and is now said to be in urgent need of updating (Bywaters 2013).

This debate has been widened by robust new evidence which suggests that it is income inequality rather than poverty which is a more powerful link with maltreatment. A large-scale US epidemiological study has shown that the effect of inequality remained significant after adjustments were made for county-level variations in child poverty and for state variations in child maltreatment rates (Eckenrode et al. 2014). The impact of income inequality was also greatest in areas with the highest child poverty rates (Eckenrode et al. 2014:459). While it is commonly argued that more equal societies and communities have fewer health and social problems than those that are less equal (Wilkinson and Pickett 2007; Wolfe et al. 2014) lower rates of maltreatment should now be added to the list of benefits of a more equal society.

Child welfare professionals in the UK report seeing more cases of suspected neglect than a year ago and attribute this to higher levels of localised poverty (Burgess et al. 2014). These anecdotal practitioner reports echo substantial evidence to suggest a strong association between poverty and child maltreatment and especially neglect. Neglect is more commonly linked with poverty than other forms of child abuse (Connell-Carrick 2003; Connell-Carrick and Scannapieco 2006; Pecora et al. 2012). The close relationship between poverty and neglect arises because a number of the factors that underlie poverty are also associated with neglect. Poverty is typically associated with unemployment, low rates of pay, being a lone parent and having a large family (Connell-Carrick 2003). Parents with a low income are more likely to feel chronically stressed than parents with higher incomes and this is exacerbated for those living in poorer neighbourhoods (Ghate and Hazel 2002).

Although poverty is a risk factor for neglect it is important to note that the majority of poor families do not neglect their children. But the increased stress which poverty creates, makes coping with the psychological as well as the physical and material demands of parenting much harder (Howe 2005; Crittenden 2008). In this respect, poverty can be one of a succession of negative consequences, which, if not halted, can add to the likelihood of poorer or even dangerous parenting.

Neglect can be defined from the perspective of a child’s right not to be subject to inhuman or degrading treatment (European Convention on Human Rights, Article 3). It has been argued that this focuses, unhelpfully, on parental intentions, when one of the distinguishing features of neglect is the omission of specific behaviours rather than the deliberate commission of abusive acts (see Connell-Carrick 2003). Defining neglect in terms of the impact on the child (likelihood of significant harm or impairment to the child’s development) puts the emphasis, instead, on whether a child’s needs are being met, regardless of parental culpability, and is the approach adopted in England. Neglect is defined in statutory guidance as:

The persistent failure to meet a child’s basic physical and/or psychological needs, likely to result in the serious impairment of the child’s health or development. Neglect may occur during pregnancy as a result of maternal substance abuse. Once a child is born, neglect may involve a parent or carer failing to:

It may also include neglect of, or unresponsiveness to, a child’s basic emotional needs (HM Government 2013, p. 85).

The ‘persistent failure to meet the child’s needs’ in the definition puts the emphasis on the frequency, enormity and pervasiveness of the behaviours that would identify them as abusive. This helps to offer a benchmark for when neglect becomes maltreatment.

There are tensions in working with this definition where it abuts family poverty. Although few parents will physically or sexually abuse their children, most will, on occasion, neglect them, not least because of force of circumstances. This makes neglect more complex, morally and socially, to identify or ‘name’ because it may be closer to normative parenting behaviour. The UK children’s charity Action for Children notes that it takes extraordinary levels of organisation and determination to parent effectively in situations of poor housing, meagre income, lack of local resources and limited educational and employment prospects (Burgess et al. 2014).

International human rights legislation demands that Nation states undertake enquiries into the deaths of children where maltreatment is known or suspected to see what more could have been done by the state (Rose and Barnes 2008). In the four UK nations these enquiries take place at a local level and the learning from these local reviews is collated nationally. Earlier national studies of small numbers of reviews found high levels of poverty among families where a child had died or been seriously injured (Brandon et al. 2002). Gauging the extent of poverty and hardship in more recent reviews is hampered because poverty is only patchily recorded (Devaney et al. 2013; Brandon et al. 2008). There are two possible inter-connected reasons for this; firstly agencies seeking to learn lessons about their own practices are not primarily concerned with improving the socio-economic circumstances of families; secondly those writing reviews may not consider this information relevant to the purpose of the review so may not record it, even though poverty is a known risk factor for maltreatment (Devaney et al. 2013; Vincent and Petch 2012). ‘Poor living conditions’ was used as the best proxy measure for poverty in a national study of reviews for England (Brandon et al. 2008) but even this information was sometimes found to be absent in a Northern Ireland study (Devaney et al. 2013).

Up to date findings about poverty rates from these reviews are therefore unreliable and likely to be an under-estimate. Hence evidence of poverty in families where children die or are seriously harmed ranges (unreliably) from just under a fifth in Scotland (11 out of 56 cases, Vincent and Petch 2012) to just under a third in England (14 out of a sub sample of 47 cases, Brandon et al. 2008) to just under a half in Northern Ireland (10 out of a total of 24 cases, Devaney et al. 2013).

The study reported here sought to learn more about the extent and meaning of neglect in the most serious maltreatment cases in England through a re-analysis of a sub sample of cases drawn from over 800 local reviews (Serious Case Reviews or SCRs) from 2003 to 2011 (Brandon et al. 2013). The wider study offered both a quantitative analysis of all cases (n = 101) where neglect had been clearly substantiated (child was named on a child protection plan for neglect) as well as a qualitative analysis of material drawn from summarised overview reports from a total of 46 available SCRs, selected because of the outcome of catastrophic neglect (see also Brandon et al. 2014). The focus here is a re-analysis of these 46 cases. Each case was analysed repeating the ecological transactional approach (Cicchetti and Valentino 2006) used in our previous SCR studies (for example Brandon et al. 2008, 2012).

The ecological transactional approach helps to understand the impact of poverty alongside other risks for neglect. It provides a theoretical framework for thinking about the dynamics between children, carers and helping agencies and the way that different risks of harm combine and interact to influence children’s development and safety. This model offers a framework for the analysis of both research and practice (Connell-Carrick 2003). It extends learning from ecological theory to take into account developmental psychopathology (for example Cicchetti and Valentino 2006). It frames the understanding of parenting capacity primarily in terms of the caregivers’ psychological sensitivity and availability to their child. Thus a major predictor of poor parenting is a lack of parental understanding of the psychological complexity of children, especially babies (Sroufe et al. 2005).

Maltreating parents’ complex patterns of behaviour and responses are in part derived from their own past experiences of relationships. Parents’ resources and ability to keep children safe and well are challenged by social and economic factors like poverty and community violence and other hardships which affect their capacity to be attuned and sensitive to their developing child.

During the analysis, a sixfold typology emerged of circumstances linked to the catastrophic neglect (deprivational neglect, medical neglect, accidents with elements of forewarning, sudden unexpected deaths in infancy, physical abuse combined with neglect and young suicide). The extent to which poverty might have played a part is considered as one of many interacting risk factors. The Joseph Rowntree Foundation definition of poverty was used to determine where families were living in poverty (Parekh et al. 2010).