Holistic,expert and centralised models for organising care placement decisions: findings from Finland

Abstract

This article examines the organisational models of decision-making regarding care placements in child welfare. Many countries currently report reorganising and economising their services and greater use of independent and voluntary agencies, all of which will have a direct and indirect influence on placement decisions about children in care. Very little is known about how public authorities make these decisions, what processes achieve the best option for a child and how different organisational models interact with front-line decision-making. The article is based on a study of placement decisions in Finland. The data comprise 49 phone interviews and five focus group interviews with social workers and their managers in charge of substitute care in different municipalities. The analysis presents three decision-making models: the holistic model, the shared expertise model and the centralised model. Each of these rests on different understandings of knowledge, expertise and client participation. The study predicts that the centralised model is likely to become more common due to the reorganisation of social services. While this may strengthen the managerial process and enhance the quality of decision-making, it is likely to diminish the relationship-based casework that is typical for the holistic model.

Keywords

Decision-making placement decisions removals of children professional discretion human service organisations

Introduction

When a child is taken into care and separated from their parents, important and far-reaching decisions have to be made, including the choice of substitute home. What kind of foster family would best serve the needs of the child? Is kinship or residential care preferable? When matching children to placements, social workers consider a number of issues, such as the child’s and parents’ needs, rights and wishes, the nature and form of the substitute home, professional expectations, legal requirements and safeguarding issues. The final choice is the result of complex judgements and decisions, often overshadowed by uncertainty about the outcome (Pösö and Laakso, 2014; Tilbury, 2007.) However, little is known about how these practices are organised and arranged to help practitioners make the best possible decisions.

Yet the organisation of decision-making practices has become more important as the reorganisation and economisation of welfare services and increasing privatisation of substitute care have influenced and reformulated its provision, as has been reported in the UK, Sweden and Finland (Höjer and Forkby, 2011; Laakso, 2013; Sellick, 2006, 2011, 2013; Stanley, et al., 2013; Wiklund, 2005). Competitive bidding by service providers – reflecting wider changes introduced by new public management philosophies – also plays its part in influencing decisions. However, the relationship between organisational contexts and front-line practice has been little studied, despite its significance for the services and children in question (Glisson, Green and Williams, 2012).

In this article we examine the ways in which the statutory decision-making of substitute care is organised in municipalities in Finland and the kind of influences that different organisational models have on the front-line practice of social workers. We employ the notion of human service technology developed by Hasenfeld (1983) to address the interconnectedness of organisational models as well as the rationales of social work decision-making. Our interest in this approach comes from observations in our other studies suggesting that placement decisions by Finnish social workers are organised differently in different municipalities and that decision-making practices also vary as a result.

The study is based on telephone interviews undertaken with social workers and managers in 49 municipalities in 2012 and on five focus group interviews with 18 social workers. Child welfare policy documents in 42 municipalities were also analysed. But before proceeding to the analysis, we will briefly describe the context of Finnish child welfare services.

The context for placement decision-making in Finland

In Finland it is the duty of municipalities to provide child welfare services, including out-of-home care, to a national population of 5.4 million. The 336 municipalities are therefore responsible for placing children in care. The service is highly decentralised in that although the municipalities have to abide by the Child Welfare Act (417/2007), they exercise considerable autonomy to fit the legislative norms to their local culture, politics and economy. Consequently, variation among the provision of public services is so wide that some researchers argue that when discussing health and welfare services, the focus should be on the local rather than the national setting (Burau and Kröger, 2004). The parlance employed to reflect the role of the municipalities is ‘local governance’. As some municipalities are very small and find it difficult to provide good-quality services due to a lack of qualified staff or finance, they have started to co-operate with others to create ‘inter-municipal’ coalitions for service provision and consultation. Such co-operation supports statutory service provision, including child welfare, as it enables relevant expertise, experience and knowledge to be shared. Some municipalities have gone even further and introduced inter-municipal agencies for placement decision-making, as will be shown later. In social and health care policy, there is a growing belief that services can only be effective if organised by large municipalities or coalitions of smaller ones (e.g. Sosiaali- ja terveysministeriö, 2013a).

In addition to local variations in service provision, as front-line decision-makers social workers have wide discretion since the Child Welfare Act and accompanying regulations only set out a general framework for interventions (Berrick, et al., 2015; Pösö, 2011). A study of the organisational cultures and climates of Finnish municipal social welfare (Rostila, et al., 2011) highlighted the considerable differences among municipalities in how far bureaucratic and professional norms and principles guided policy and practice. Most importantly, the study also recognised differences within the municipalities themselves. This is to suggest that it is not only local politics, culture and traditions that matter but also the smaller work units within them. In such a decentralised system, it is not surprising that there have been calls for more uniform service provision, practice and eligibility criteria (e.g. Sosiaali- ja terveysminsteriö, 2013b).

The selection of a substitute home for a child entering care is made by social workers employed by the country’s statutory social welfare agencies. If the child is 12 years of age or older and his or her custodians agree with the choice, the child will move to the new home. If there is any disagreement, the final decision is made by the administrative court. As is common in many countries, all decisions have to be informed by the principle of the child’s best interests being paramount, with particular attention paid to his or her needs, wishes, family contacts and linguistic, cultural and religious background (Child Welfare Act 417/2007, section 50). In practice, the needs of children taken into care vary considerably as the scope of the Finnish child welfare system is wide, dealing with abuse, neglect and delinquency, as is typical of other Nordic countries (Pösö, et al., 2013).

After the decision to take the child into care has been made, the placement options recognised by the present Child Welfare Act (417/2007) comprise family foster homes or residential institutions or ‘some other way required by the child’s needs’. The option of kinship care has to be examined in every case, although it currently only applies to nine per cent of children in foster care (Laakso, 2013). In addition, although legislation requires that placement in foster care should always be prioritised and its potentiality examined first, only 49% of children are so placed. However, rates vary across municipalities: in 2011, the lowest figure in a municipality was 39% and the highest 62% (Laakso, 2013: 17).

The wide spectrum of children’s needs, rising numbers, social work discretion and trends towards service reorganisation and decentralisation produce a specific context for child welfare decision-making in Finland. From the municipalities’ perspective, more out-of-home placements and more generous staffing have raised the costs of providing good-quality care and budgets and value for money have become crucial policy issues. The absence of any systematic analysis of placement decision-making practices and organisational arrangements is, therefore, a serious deficiency in the information needed to fashion an informed policy (Laakso, 2013).

The conceptual agenda

Human service technology is the concept used to analyse the ways in which municipalities, as the key public administration bodies, organise placement decision-making when a child enters care. This approach helps us to move from individual cases to processes and systems – a concept defined by Hasenfeld (1983: 111) as:

… a set of institutionalised procedures in human service organisations aimed at changing the physical, psychological, social, or cultural attributes of people in order to transform them from a given status to a new prescribed status.

The term ‘institutionalised’ denotes that the procedures are legitimated and sanctioned by the organisation. As it is front-line social workers who follow the mandated procedures, they are the key players in applying the technology.

A special feature of human services is the fact that they are about, with and for people (Hasenfeld, 1983: 115–116). Therefore human service technologies have to cope with the uncertainties related to the variability and complexity of individuals as well as to the limited understanding of the relationship between the intervention and its outcomes (Hasenfeld, 1983: 114). In fact, a degree of uncertainty about the outcomes of decisions is seen as a fundamental characteristic of social work decision-making (e.g. O’Sullivan, 2011; Munro, 2008). In the case of placement decisions, even well-planned arrangements can falter if the child runs away, the foster carers experience a sudden change, such as divorce, or privately run residential homes go bankrupt (Pösö and Laakso, 2014).

According to Hasenfeld (1983: 125–134), human service technologies comprise different components each of which focuses on:

the characteristics of the clients – client attributes technology;

the knowledge used in the service delivery – knowledge technology;

interaction patterns – interaction technology;

the control of clients to maximise the intended effects – client control technology;

the organisations of staff activities – operations technology.

In child welfare, especially in statutory decision-making, these components highlight the spectrum of issues that social workers have to consider and the range of activities they undertake. The child’s and his or her family’s characteristics are examined (client attributes technology), knowledge is acquired, tested and documented for preparing and making decisions (knowledge technology), different groups of people meet to make decisions and options are discussed face to face and recorded in case files (interaction technology), the suitability of the children as well as the implications of interventions are assessed and reported (client control technology) and the package of organisational tasks is shared by different members of staff (operations technology).

In addition, operations technology includes the actual availability and use of substitute homes. Practitioners stress that the shortage of beds in substitute homes is one of the key issues that challenge placement decisions in individual cases (e.g. Vettenranta, Holma and Rousu, 2008). As this is dependent on the willingness of ‘private’ families to become foster carers, the ability of independent agencies to provide residential places (increasingly so in Finland – see Bardy and Heino, 2013) and the capacity of the municipality to fund what is needed, the human service technology in child welfare is influenced by factors external to the organisation itself.

Data and method

Our analysis had two aims: first, we were interested in how the decision-making was organised (the structure for human service technology, in other words the organisational model) in different municipalities; and second, we wished to chart the range of knowledge, client assessment and participation and organisational interactions (client attributes, knowledge, interaction, client control and operations technology) used in each organisational model. The analysis is based on empirical data gathered in 2012 in Finland.

The data consist of two types of interview material and written policy documents. First, a geographically representative sample of 57 municipalities out of a total of 336 was drawn up so that it included municipalities with different care profiles (e.g. the rate of children in care and the use of foster and residential care) and with different arrangements for service provision (e.g. inter-municipal coalitions, sole municipal child welfare). To achieve this, information was extracted from national child welfare statistics and material provided by the Centre of Excellence on Social Welfare, which has specialist knowledge of the models operating in Finland.

Following a request to participate, 49 phone interviews were carried out with team leaders and social workers making decisions about children in care. The key questions focused on how the placement decision-making was organised and how the present system worked. The interviews were tape-recorded. Next, five agency-based focus group discussions were held with 18 practitioners comprising experienced social workers and their team leaders from a number of municipalities. The focus here was more on the role of decision-making in casework. Each group organised itself independently, as is recommended in the methodological literature on arranging focus group interviews (Bloor, et al., 2001; Morgan, 2002).

Finally, policy documents specifying the welfare plans for children and young people were collected from the participating municipalities and relevant central government departments. These plans are required by the Child Welfare Act (417/2007) and should present the general state of children’s and young people’s welfare in the municipality and the concrete steps taken to improve it. Forty-two plans were studied in order to establish the range of policy guidelines and the decision-making processes they supported. Surprisingly, we found that they included little about substitute care at a general level and nothing at all about organising placement decision-making.

In this article, we focus on the organisational models which the municipalities have introduced to make placement decisions. The fact that many municipalities separate decision-making on residential care from that on foster care complicates the typification of the models: the holistic model, the shared expertise model and the centralised model. These are constructed on the basis of information about the organisation that makes the decisions, the practitioners involved, the process and the nature of knowledge required.

The second part of our analysis looks at the rationales behind the models and how and why they are used in social work decision-making.

Organisational models for matching

The holistic model

The holistic model rests on the caseworkers’ knowledge and judgement of the needs of the child and of the quality and suitability of the placement for meeting them. In this model, placement decisions form an integral part of casework. Social workers may consult others to gain information and support, but it is they who are in charge and responsible for weighing up the options.

As the social worker has often been working with the child before the need for substitute care, she or he has come to know the child, their family and circumstances, and thus the assessment of the child’s needs incorporates ‘holistic’ knowledge of the child’s and family’s situation as well as their wishes.

In this model, it is the social worker’s responsibility to be well informed about the different options and the profiles, quality and availability of possible settings. It is also necessary to be aware of the competitive bidding for different options. Thus, the term ‘holistic' not only covers knowledge of the child but also of the service provision.

The shared expertise model

The model of shared expertise (expert group) rests on the assumption that the placement decision requires the knowledge and expertise of different professionals working together. In other words, multi-professionalism is regarded as essential. In larger municipalities, this group assembles practitioners from different fields of child welfare, such as the managers of children’s homes or practitioners working with foster parents, and shares their knowledge. The group may serve different municipalities and members can come from any of them. The expertise is both practical and psychosocial. Although the group has a formal position in the decision-making system and is involved at every stage, its role is consultative and the final choice remains with the caseworker.

The centralised model

In the centralised model, a group of professionals from one or several municipalities collects information about the quality and availability of placements, manages the competitive bidding process and uses this knowledge to suggest a suitable home for a child. Their expertise mainly covers the available provision and less attention is paid to the needs of the child. The approach is based on the assumption that the service provision is so broad and uneven that such a group is needed for quality control. Hence, one of its tasks is quality control. Caseworkers seeking a placement for a child have to present their assessment to the group, usually via a written report. Naturally, the caseworker may disagree with the decision, but the underlying assumption is that this group has the most up-to-date and relevant understanding of what is appropriate and its proposals should be seriously considered.

The rationales of the models

We have seen that the holistic model perceives the knowledge required for placement decisions as an integral part of general casework whereas the other two regard it as a more distinctive area of expertise. The centralised model goes furthest in terms of fragmentation of expertise, resting on the notion that matching child and placement requires very specialised knowledge and information. This expertise is owned by the group of specialists and caseworkers are required to ask for instructions and advice. It views the necessary expertise in terms of service provision management in contrast to the expert groups model that sees it more in terms of practical knowledge. Subsequently, the models have incorporated different knowledge technologies (Hasenfeld, 1983). They all implement the same legal principles and norms set by the Child Welfare Act, but they have built different understandings of what kind of knowledge is required to provide an effective service.

The models also differ in the ways they accumulate information on the child in question. The holistic model is based on the relationship between the caseworker and the child, whereas the other two models rely on representations of the children by their caseworkers and the information they provide. Following Hasenfeld’s terminology, the client attributes technology is different in each model. Furthermore, the models of experts and centralisation assume that interaction between the professionals is necessary, whereas the holistic model rests on the interaction between the social worker and client. In addition, the expert group and the centralised models deal with every placement in the municipality in order to standardise the matching process, whereas the holistic model allows more child-centred discretion.

Table 1 summarises the rationales of each model.

According to the interviews with practitioners, the holistic model functioned well when social workers had long experience of local child welfare. They would know the children and families and what provision was available locally and regionally. Moreover, they tended to establish professional networks that they could consult informally when needed, making use of the collective memory of practitioners (e.g. Forkby and Höjer, 2011).

The characteristics of the holistic model are demonstrated by the following extracts from a phone interview with a social worker and a group interview:

I mean, the social worker … usually knows the child’s needs best, they can sort of evaluate if a certain place corresponds to the needs of this child and if it can secure all those things that you have to secure. (Social worker, telephone interview)

For me, it almost means the most that I know the place, that I have experience with it. That maybe I’ve placed someone or someone else from our team has placed someone there … I can talk on the phone with a colleague even in another town, unofficially, about what kind of an impression I got from visiting an institution, or what kind of experiences I have from there. (Social worker, group interview 1)

The rationale behind the model of the expert groups is to provide specialisation in terms of expert knowledge. The assumption is that the child’s caseworker does not have enough appreciation of all the nuances relevant to making a placement decision. In the interviews, the practitioners highlighted the fact that expertise is required to understand all the varieties, philosophies and strengths of the possible options. They claimed that it was impossible for an individual social worker to master all this information. The following extract describes a situation where the knowledge of the group surpassed that of the social worker:

We obviously have examples of when you’ve had a really strong idea that you don’t have any other option than an institution, but then you get a message from somewhere that you might find a foster family and change direction. And that’s where I think shared discussion is very important – to make sure that you take into account all possible perspectives. But the basic setting is that the child’s needs take priority and we can then scan the places we know to see if they match at all or not. (Social worker in charge of substitute care, group interview 5)

The centralised model serves municipalities that place a high number of children who need a wide array of care settings. The management of service provision is seen as a necessary part of child welfare as it is so specialised and extensive. Since management is the key element in this model, direct knowledge of the child is largely irrelevant to the matching process; the caseworker’s assessment report on the child and family is adequate for gathering the necessary information. Moreover, an indirect relationship with the child can help to avoid emotional attachments that might blur professional judgement on what is required. So, instead of emotionally argued choices, this model favours fact-based standardisation as the basis for effective decision-making.

Table 1.

The rationales of the three organisational models.

Human service technology in use	Holistic model	The model of experts	The centralised model
Client attributes technology	Direct face-to-face casework practice	Indirect presentation of children	Indirect and standardised presentation of children
Knowledge technology	Child-based discretion of casework	Specialisation: Multi-agency and multi-professional knowledge of substitute care	Specialisation: extensive knowledge of substitute care provision
Interaction technology	Interaction between the caseworker, child and the family	Inter-professional and inter-agency interaction	Inter-professional interaction
Client control technology	Negotiation between the caseworker, child and family	Consultation with the caseworker	Standardisation of decisions
Operations technology	Relationship-based casework	Shared practical knowledge of substitute care, standardisation	Management of the placements through standardisation

The rationale behind the centralised model is presented in the following extracts:

I mean, you have quite clear instructions here that a social worker should not comment on the place too strongly. Of course they can express their view about the type of services needed through their presentation of the child’s needs, but to name a place or say something like ‘it has to be outside their own town’ is unacceptable. Instead they should describe the child’s needs; it is then our task to find a suitable place for the child based on the profile of needs. This is the instruction. (Leader of centralised placement unit, group interview 4)

Last spring, we produced written service descriptions of the different forms of substitute care. So a service description means that it has the service described in detail, especially what it includes, such as how many hours a week a child needs that activity in that type of service. It covers specialised provision, residential care, family rehabilitation, foster care, and so on, and then seeks to show which of the child’s needs it can meet best. (Leader of centralised placement unit, group interview 4)

Each of the models has its strengths and weaknesses. The holistic model is vulnerable if the social workers and/or the service provision change a lot; it requires stability in the social work community, a situation that is unfortunately becoming less common in Finland due to staff turnover and changes in services. The decisions may also be very personal to the professional.

In contrast, the model of expert groups is dependent on shared practical knowledge which, for similar reasons, is subject to changes, whereas the centralised model risks losing the child’s perspective by prioritising the need for strong management. The last two models have considerable weaknesses from the point of view of children’s participation as important steps in the process exclude children’s views or attendance; the emphasis is on inter-professional co-operation rather than client participation. This is an important feature to consider in Finland where, on the one hand, child participation is high on the agenda of the legislation and, on the other, where the majority of children taken into care are teenagers who, due to their age and development, are especially entitled to participate in decisions affecting them.

Discussion and conclusions

The variation in organisational models and human service technology underpinning placement decisions demonstrates the extent of ‘local governance’ in Finland (cf. Brau and Kröger, 2004).

Nevertheless, regardless of the organisational model, the practitioners informed us that from their point of view matching practices in their area functioned well. According to them, their decision-making practices adhered to the key principle of the child’s best interests being paramount (Laakso, 2013). They did acknowledge that sometimes decisions were compromised, meaning that the child had to enter a home that was only second or third choice, but these difficulties were due to factors such as the unavailability of places, unexpected events in the substitute homes or changes in the child’s and their parents’ wishes rather than to the prevailing organisational model (Pösö and Laakso, 2014). Hence, professionals appeared to accept what they found in their municipality and maintain this practice. This satisfaction with the status quo might reflect loyalty towards their organisations or attempts to provide politically correct answers, but equally could be an objective and well-grounded assessment of the quality of the decision-making in their area.

Our data are too limited and selective to allow us to conclude whether the organisational models influence the quality of the decisions. More research is needed to explore this and how it relates to the organisational models employed. Research designs that examine the decision-making practices in situ and follow up placements, ascertaining the views of practitioners, children, parents and substitute caregivers, would be especially important.

Considering the present policy emphasis on large inter-municipal units in health and social welfare in Finland, the holistic model, resting on the principles of small and stable communities and related knowledge, is likely to be replaced by different models, especially the centralised one. This development will emphasise specialisation and might well increase the fragmentation of personal knowledge because it distances the crucial elements of placement choice from casework practice and involves the children only indirectly. The management of service provision would come to dominate at the expense of sensitivity to children’s views and social relationships. Similar concerns have been raised about social welfare in general due to the ways it is being managerialised (e.g. Garrett, 2009; Harris and White, 2009). However, studies in Sweden (Forkby and Höjer, 2011) and Finland (Laakso, 2013) have shown that even in the era of new public management, professionals may find a way to circumvent restrictive administrative regulations if the child’s best interests and professional judgement so suggest.

The likelihood of the managerial view of service provision taking over child-related views is especially alarming as in Finland a growing number of care services are being farmed out by public authorities to private agencies (Bardy and Heino, 2013) and more time and attention are devoted to quality audits and costing exercises. This is presenting an obvious challenge to keep the focus on the needs and rights of children requiring a substitute home and to make sure that local prevailing practices are not accepted blindly (e.g. Forkby and Höjer, 2011; Sellick, 2013; Stanley, et al., 2013). Therefore, the message from Hasenfeld (1983) about human service technology being about/for/with human beings, is important to revisit; organisational models are not neutral management frameworks but reflect significant and contentious political and ethical standpoints about what is important in child welfare.

Footnotes

Tarja Pösö is a Professor in the School of Social Sciences and Humanities, University of Tampere, Finland.

Riitta Laakso, PhD is a Researcher in the same department.

References

Bardy

Heino

(2013) Katsaus lastensuojelun toimintaympäristöihin: paniikista toivoon ja näköalat auki [An overview of the contexts of child welfare: from panic to hope]. In: Bardy

(eds) Lastensuojelun ytimissä [In the Core of Child Welfare], Helsinki: THL, pp. 13–42.

Berrick J, Peckover S, Pösö T and Skivenes M (2015) The formalized framework for decision-making in child protection care orders: a cross-country analysis. Journal of European Social Policy.

Bloor

Frankland

Thomas

Robson

(2001) Focus Groups in Social Research, London, Thousand Oaks and New Delhi: SAGE Publications.

Burau

Kröger

(2004) The local and the national community care: exploring policy and politics in Finland and Britain. Social Policy & Administration 38(7): 793–810.

Child Welfare Act (2007/417) Available at: www.finlex.fi/en/laki/kaannokset/2007/en20070417.

Egelund

(1996) Bureaucracy or professionalism? The work tools of child protection services. International Journal of Social Welfare 3(5): 165–174.

Forkby

Höjer

(2011) Navigation between regulations and gut instinct: the unveiling of collective memory in decision-making processes where teenagers are placed in residential care. Child & Family Social Work 16(2): 159–168.

Garrett

(2009) ‘Transforming’ Children’s Services? Social work, neoliberalism and the ‘modern’ world, Maidenhead: Open University Press.

Glisson

Green

Williams

(2012) Assessing the Organizational Social Context (OSC) of child welfare systems: implications for research and practice. Child Abuse & Neglect 36(9): 621–632.

10.

Gubrium

Holstein

(1997) The New Language of Qualitative Methods, New York: Oxford University Press.

11.

Harris

White

(2009) Modernising Social Work: Critical considerations, Bristol: The Policy Press.

12.

Hasenfeld

(1983) Human Service Organizations, Englewood Cliffs, NJ: Prentice-Hall.

13.

Heino

(2009) Lastensuojelu – kehityskulkuja ja paikannuksia [Child protection – tendencies and locations]. In: Lammi-Taskula

Karvonen

Ahlström

(eds) Lapsiperheiden Hyvinvointi [The Well-being of Families with Children], Helsinki: Terveyden ja hyvinvoinnin laitos, pp. 198–213.

14.

Höjer

Forkby

(2011) Care for sale: the influence of new public management in child protection in Sweden. British Journal of Social Work 41(1): 93–101.

15.

Laakso

(2013) Mistä koti huostaan otetulle lapselle? [Where to Find a Home for a Child who has been Taken into Care?], Helsinki: Kunnallisalan kehittämissäätiö.

16.

Munro

(2008) Effective Child Protection, London: SAGE Publications.

17.

Morgan

(2002) Focus group interviewing. In: Gubrium

Holstein

(eds) Handbook of Interview Research, London, Thousand Oaks and New Delhi: SAGE Publications, pp. 141–160.

18.

O’Sullivan

(2011) Decision-making in Social Work, Basingstoke: Palgrave Macmillan.

19.

Pösö

(2011) Combatting child abuse in Finland: from family to child-centered orientation. In: Gilbert

Parton

Skivenes

(eds) Child Protection Systems: International trends and orientations, New York: Oxford University Press, pp. 112–130.

20.

Pösö

Laakso

(2014) Matching children and substitute homes: some theoretical and empirical notions. Child & Family Social Work. doi:10.1111/cfs.12144.

21.

Pösö

Skivenes

Hestbaeck

(2013) Child protection systems within the Danish, Finnish and Norwegian welfare states: time for a child centric approach? European Journal of Social Work. doi:10.1080/13691457.2013.829802.

22.

Rostila

Mäntysaari

Suominen

Asikainen

(2011) Sosiaalityön professionaalisuuden erot kuntaorganisaatioissa – Organisaatiokulttuurien ja organisaatioilmapiirien kirjo [The differences between social work professionalism in the municipal organisations – the variation of organisational cultures and atmospheres]. Janus 19(2): 143–157.

23.

Stanley

Austerberry

Bilson

Farrelly

Hussein

Larkins

Manthorpe

Ridley

(2013) Turning away from the public sector in children’s out-of-home care: an English experiment. Children and Youth Services Review 35(1): 33–39.

24.

Sellick

(2006) Opportunities and risks: models of good practice in commissioning foster care. British Journal of Social Work 36(8): 1345–1359.

25.

Sellick

(2011) Privatising foster care: the UK experience within an international context. Social Policy & Administration 45(7): 788–805.

26.

Sellick C (2013) Foster care commissioning in an age of austerity: the experiences and views of the independent provider sector in one English region. British Journal of Social Work. Advance Access published March 13. doi:10.1093/bjsw/bct046.

27.

Sosiaali- ja terveysministeriö (2013a) Sosiaali- ja terveydenhuollon järjestämislain valmisteluryhmän loppuraportti. Report 45, Helsinki: Sosiaali- ja terveysministeriö.

28.

Sosiaali- ja terveysministeriö (2013b) Toimiva lastensuojelu [Functioning child welfare] Report 19. Available at: www.stm.fi/c/document_library/get_file?folderId=6511574&name=DLFE-26809.pdf.

29.

Tilbury

(2007) The regulation of out-of-home care. British Journal of Social Work 37(2): 209–224.

30.

Vettenranta

Holma

Rousu

(2008) Lastensuojelupalvelujen hankkimisen ja tuottamisen näköaloja [The Landscape for the Purchasing and Providing of Child Welfare Services], Helsinki: Suomen Kuntaliitto.

31.

Wiklund

(2005) Social barnavård i nya former. Om marknadsorientering inom barnavårdsarbetet [New forms of social child welfare: about the market orientation in child welfare work]. Socialvetenskaplig tidskrift 2–3: 149–168.