Is child protection an inequalities issue Paul Bywaters

Is child protection an inequalities issue? Paul Bywaters Coventry University

Structure 1. Assumptions and arguments 2. Key findings of West Midlands study 3. Introduction to 4 nations study Throughout trying to examine the Warwickshire data and pose key questions. No easy answers. No clear answers from me!

How does research come about?

Practice ‘Variations’ and Poverty • Poverty is repeatedly identified by research as biggest single factor producing ‘variations’ in child welfare intervention rates but little detailed research on this relationship. • A failure to conceptualise variations as inequalities, as an issue of social justice, either by researchers, policy makers or practitioners – unlike for health and education policy and practice. ‘The … socio-economic context in which parents are expected to raise their children is easily ignored. ’ Bradt et al 2014.

Child welfare: an issue of social inequity Definition: Child welfare inequity occurs when children and/or their parents face unequal chances, experiences or outcomes of involvement with child welfare services that are systematically associated with structural social disadvantage and are unjust and avoidable.

Child welfare: an issue of social inequity Inequity in child welfare chances: Who receives child welfare interventions? Inequity in child welfare experiences: Which children/families get what kinds of interventions? Inequity in outcomes: What differences are there in outcomes between children involved with child welfare services and those who are not?

Child welfare inequities: health inequalities 4 key concepts: • The social determinants of health: the conditions in which we are born, grow, live, work and age. These conditions are shaped by inequalities in power, money and resources, and therefore are unequal in their distribution. This inequality in the social determinants of health contributes to inequalities in health outcomes.

Child welfare inequities: health inequalities 4 key concepts: • The social gradient in health ‘Within any particular society, those with higher incomes do better on a range of outcomes. There is therefore a ‘social gradient’ in health, which means that every step up the socio-economic ladder leads to an increase in health. ’ Karen Rowlingson 2011 ‘Does income inequality cause health and social problems? ’ Joseph Rowntree Foundation The steeper the gradient the greater the impact of socio-economic factors on health. It is not only those at the bottom of the ladder who suffer poor health.

Child welfare inequities: health inequalities 4 key concepts: • Intersectionality ‘People’s experience is not shaped by one aspect of their identity alone but by a combination of elements. Gender, age, religion, disability, health, location and migration history can all be as important as ethnicity. ’ (Barnard & Turner 2011, p. 4)

Child welfare inequities: health inequalities 4 key concepts: • Inverse care law ‘ ‘that the availability of good medical care tends to vary inversely with the need of the population served. ’ Tudor Hart, 1971, p. 411) Less affluent populations with greater health needs, tend to receive less good services than more affluent populations with less health needs.

The Coventry Study: Aims • To what extent are children’s services ‘clients’ clustered in areas of greatest deprivation? • How do gender, age, ethnicity and care status add to our understanding of this relationship? • Do patterns of deprivation within LAs explain more of the differences between LAs in children’s services’ outcomes than overall IMD scores? • Is deprivation a more powerful factor in explaining inequalities in LAC rates than inequalities in rates for children in need or on child protection plans?

The Coventry Study: Methods Sampling: 13 Local Authorities in the English West Midlands. All ‘children in need’, which includes all children in out of home care (Looked after children or LAC) and children on child protection plans (CPP) Over 10% of all children England of CPP and LAC but a convenience not a representative sample.

The Coventry Study: Methods Data Collected Routinely reported data for all CPP and LAC: age, gender, ethnicity, disability, reason for CPP and legal status in LAC at 31. 3. 12. Used national guidance for definitions etc. + Neighbourhood (Lower Layer Super Output Area: LSOA) of family origin. Population data from the 2011 Census Index of Multiple Deprivation scores for LAs and LSOAs

The Coventry Study: data issues Data cleaning: • Able to check against published data but not a perfect match • Some data absent: unknown birth dates; Unaccompanied Asylum Seekers • Some data conceptually dubious: ethnicity • Some data collection poor and inconsistent: disability No data on individual family circumstances but proxy No data about the characteristics of neighbourhoods beyond deprivation scores and demography

The Coventry Study: Analysis Sorted neighbourhoods (LSOAs. MSOAs) by their national deprivation rank using 2010 Index of Multiple Deprivation scores and divided into deciles (10 groups of 10%) or quintiles (5 groups of 20%). Where we refer to decile 10 in our sample, it means those neighbourhoods in the Midlands sample that are in the 10% most deprived LSOAs nationally. Decile 1 are the neighbourhoods amongst the least deprived 10% in England not the most deprived in the Midlands.

Deprivation Amongst Child Population Table 1: Percentage of child population living in each quintile of neighbourhoods (MSOAs) by deprivation. (Source: 2011 Census) 1 = most affluent 20% of neighbourhoods; 5 = least affluent. Deprivation Quintile 1 Child population in England Child Population Midlands Sample All 0 -4 Birmingham Warwickshire Nuneaton 2 3 4 5 19. 4% 18. 3% 18. 5% 20. 1% 23. 7% 12. 1% 15. 9% 15. 7% 18. 1% 38. 2% 10. 1% 2. 3% 30. 3 11. 6% 14. 1% 3. 8% 24. 4 13. 5% 15. 1% 13. 0% 20. 5 22. 8% 18. 7% 16. 2% 16. 8 32. 0% 42. 0% 64. 7% 8. 0 20. 2%

Key Finding 1: Very Large Inequalities There are very large inequalities in children’s chances of being on a child protection plan or being a looked after child (in out of home care), systematically and significantly related to deprivation levels.

Key Finding 1: Very Large Inequalities

Key Finding 1: Very large inequalities in rates

Key Finding 1: Very large inequalities in rates Looked After Children Rates per 10, 000 Children by Deprivation Decile, Midlands and W'wickshire 200, 0 31. 3. 12 Rate 150, 0 100, 0 50, 0 1 Rate 9, 2 Warwickshire 19, 3 2 Deprivation 3 4 Deciles, 5 7 Affluent 8 9 10 16= Most 19, 0 20, 9 23, 0 29, 4 37, 4 55, 0 79, 5 94, 0 108, 0 31, 6 37, 1 24, 1 57, 0 72, 6 69, 3 77, 9 152, 0 172, 6

Key Questions 1 • • Given the very large inequity in rates of LAC related to deprivation levels, are resources allocated between areas in ways that reflect deprivation? Where would you place your effort in order to reduce demand for the most expensive services? 1 2 3 4 5 6 7 8 9 10 ALL W’shire LAC Numbers 32 55 58 28 64 85 77 60 65 81 605 LAC % of total 5 9 10 5 11 14 13 10 11 13 W’shire LAC Rates 19. 3 31. 6 37. 1 24. 1 57. 0 72. 6 69. 3 77. 9 152. 0 172. 6 54. 1 Midlands Rates 9. 2 19. 0 20. 9 23. 0 29. 4 37. 4 55. 0 79. 5 94. 0 108. 0 60. 7 Ratio W/Mids 2. 1 1. 7 1. 8 1. 0 1. 9 1. 3 1. 0 1. 6 0. 9

Key Finding 2: A Gradient of Inequality There is a gradient in rates across levels of deprivation, just as there is a gradient in other outcomes (health, education) for children across the whole of society. Every step increase in deprivation brings a step increase in intervention rates. Deprivation is a key factor but CPP and LAC are not found only in areas of high deprivation. Overall 60% of CPP and LAC live in the most deprived 20% of neighbourhoods but 40% live in more affluent 80% of neighbourhoods.

Key Finding 2: A Gradient of Inequality

Key Finding 2: A Gradient of Inequality Rate per 10, 000 children Child Protection Plan Rates by Deprivation Decile, Midlands Sample, 31. 3. 12 220, 0 170, 0 120, 0 70, 0 20, 0 1 Rate 6, 3 Warwickshire 13, 9 -30, 0 2 7, 5 11, 5 3 16, 2 30, 7 4 12, 9 20, 7 5 22, 1 41, 8 6 25, 3 50, 4 7 38, 2 83, 7 Deprivation Deciles, 1 = Most Affluent 8 9 10 47, 0 56, 9 68, 5 74, 0 126, 3 213, 1

Key Findings 2: A Gradient of Inequality Each step increase in deprivation has a greater effect on out of home care (LAC) rates than on child protection plan (CPP) rates. Reducing inequalities in rates between and within areas is a possible policy objective. If we could reduce the steepness of the gradient of deprivation - the impact of deprivation on family life we could reduce demands on children’s services.

Key Question 2 Could Warwickshire’s socio-economic gradient be less steep? CPP Numbers CPP Rates 1 2 3 4 5 6 7 8 9 23 20 48 24 47 59 93 57 54 10 ALL 100 525 13. 9 11. 5 30. 7 20. 7 41. 8 50. 4 83. 7 74. 0 126. 3 213. 1 46. 9 Midlands Rates 6. 3 7. 5 16. 2 12. 9 22. 1 25. 3 38. 2 47. 0 56. 9 Ratio W/Midlands 2. 2 1. 5 1. 9 1. 6 1. 9 2. 0 2. 2 1. 6 2. 2 68. 5 38. 3 3. 1 1. 2

Key Finding 3: Intersection of Ethnicity and Deprivation ‘children from black and mixed heritage backgrounds are over-represented among children who are looked after and Asian children tend to be under-represented’ (Owen and Statham 2009) ‘it is clear that minority ethnic children are over-represented in the care population’ (Selwyn and Wijedesa 2011)

Key Finding 3: Inequalities by Ethnicity Population 0 -17 by Ethnic Group in Deprivation Quintiles 4 and 5 (%) Midlands Quintile 4 5 4+5 White 19. 3 27. 8 47. 1 Mixed 18. 3 53. 3 71. 6 Asian 14. 3 67. 5 81. 8 Black 12. 7 76. 5 89. 2

Key Finding 3: Inequalities by Ethnicity

Key Finding 3: Inequality by Ethnicity

Key Questions 3: Are these ethnic differences appropriate? Warks CPP + LAC 1 2 3 White 33. 4 63. 5 52. 5 4 5 6 7 8 9 10 ALL 80. 6 117. 5 127. 1 147. 6 194. 6 400. 3 463. 4 100. 6 Mixed 0 14. 9 127. 7 238. 5 153. 1 107. 1 148. 5 301. 9 405. 4 1707. 3 153. 5 Asian 0 0. 0 Black 0 0. 0 320. 5 Other 0 0. 0 44. 0 33. 2 0. 0 235. 5 151. 5 153. 1 0. 0 107. 5 150. 0 84. 7 337. 1 0. 0 61. 3 0. 0 90. 6 0. 0 72. 5

Key Finding 4: An ‘Inverse Intervention Law’ Overall a child’s chances of an extreme child welfare intervention is much greater at higher levels of deprivation, but for a given level of deprivation a child in a more affluent local authority is more likely to be on a CPP or to be a looked after child.

Inverse intervention meets the social gradient

Child population by deprivation quintile (5 is most deprived) 1 2 3 4 5 Nuneaton and Bedworth 11. 6 13. 5 22. 8 32. 0 20. 2 Walsall 10. 0 9. 8 9. 9 15. 5 54. 8

Inverse intervention meets the social gradient

Inverse intervention meets the social gradient

Inverse intervention meets the social gradient CPP + LAC RATES NUN/BED CPP + LAC RATES WALSALL 1 2 3 4 5 62. 5 66. 7 118. 2 165. 4 355. 5 9. 5 24. 1 58. 8 76. 3 166. 3

WHY?

IIL: Explanatory Model

Key question 4 • Are higher rates or lower rates better for children? If Nuneaton and Bedworth had Walsall LAC Rates for each quintile of neighbourhood deprivation, there would have been 134 LAC at March 2012 instead of 231. If Nuneaton and Bedworth had Walsall CPP Rates for each quintile of neighbourhood deprivation, there would have been 72 CPP at March 2012 instead of 233.

Does inverse intervention apply within W’shire? CPP Walsall 1 2 3 4 5 Nun/Bed Rugby N. Warks Stratford Warwick 1. 6 27. 5 9. 2 27. 6 13. 7 6. 4 20. 0 28. 9 22. 7 18. 4 36. 1 39. 7 66. 2 81. 7 61. 6 49. 9 16. 3 70. 0 66. 8 90. 4 84. 8 55. 5 197. 1 LAC Walsall 1 2 3 4 5 Nun/Bed Rugby N. Warks Stratford Warwick 7. 9 35. 0 43. 5 25. 3 27. 5 17. 6 46. 7 36. 4 22. 7 37. 5 62. 0 19. 1 52. 0 65. 3 47. 7 66. 5 60. 0 95. 4 66. 8 113. 0 68. 5 110. 8 158. 4

Prevention? • If it is the case that the more affluent LAs are intervening more than disadvantaged LAs because of more resources relative to demand, how can that be turned into preventive early intervention without net widening?

CAB Specialist Intervention Team Model

New 4 nations UK study Identifying and Understanding Inequalities in Child Welfare Intervention Rates: comparative studies in four UK countries April 2015 – March 2017 Collaboration between 7 UK universities Funder: Nuffield Foundation Case studies in Nuneaton and Bedworth (subject to approval)

• • • Why does an inequalities perspective matter? Locates child welfare as a moral issue Points upstream – prevention as a structural issue New policy goals – reducing inequities Basis for alliances – with other fields: health, education Implications for Data collection and analysis – including family circumstances Resource allocation - taking full account of deprivation Managers – focus on neighbourhood and family deprivation Front line practice – incorporate deprivation in understanding and intervention – Education – preparing students for new practice – Research – developing understanding and testing interventions – –

For more information Web Address http: //www. coventry. ac. uk/cwip To become a stakeholder email Sophie Blackmore ac 0672@Coventry. ac. uk

For more information Thanks to the Nuffield Foundation for funding the study, to the participating local authorities and especially their information managers and to the Family Rights Group and Research in Practice for support in carrying out the project. For further details: Bywaters, P. , Brady, G. , Sparks, T. and Bos, E. (2014) ‘Child Welfare Inequalities: new evidence, further questions’, Child and Family Social Work, online access. Bywaters, P. , Brady, G. , Sparks, T. and Bos, E. (2014) ‘Inequalities in Child Welfare Intervention Rates: the intersection of deprivation and identity’, Child and Family Social Work, online access. Bywaters, P. , Brady, G. , Sparks, T. , Bos, E. , Bunting, L. , Daniel, B. , Featherstone, B. , Morris, K. , and Scourfield, J. (2015). ‘Exploring inequities in child welfare and child protection services: Explaining the “inverse intervention law”’. Children and Youth Services Review 57, 98– 105