Introduction
The Troubled Families Programme (TFP), established by the Conservative and Liberal Democrat coalition in 2012 (DCLG, 2012), is the latest example of a tradition of family intervention projects (FIPs) stretching back to the 1940s (Starkey, Reference Starkey2000). A considerable level of valuable social policy analysis of this mode of direct welfare delivery to families in the UK has been undertaken (Clarke and Hughes, Reference Clarke and Hughes2010; Sen, Reference Sen2016). This analysis has identified both negative implications and positive outcomes for families, making the overall evidence basis for the success of such programmes ambiguous (Crossley, Reference Crossley2015). However, research does suggest features such as the additional amounts of time, small caseloads and flexibility of key workers on FIP's offer advantages for service users over, for example, statutory social work services as currently organised in the UK (Broadhurst et al., Reference Broadhurst, Wastell, White, Hall, Peckover, Thompson and Davey2010). This article draws on an evaluation of an FIP, delivered by a Unitary Authority as part of the TFP between 2013 and 2015, to consider these issues. Firstly, it will critically situate the TFP within its social, historical and policy context. Secondly, the methodology employed in this study and its associated strengths and limitations will be discussed. Thirdly, the article will then draw on findings from the evaluation to consider how they relate to issues identified in previous family intervention research. Finally, the possible impact that the expansion of the TFP may have upon those involved will be discussed in light of the findings.
Context and current literature
After the summer riots of 2011 in England and Wales, the Prime Minster David Cameron claimed that responsibility for the disturbances could, in part, be apportioned to what he called ‘troubled families’ (Levitas, Reference Levitas2012). As a result, in April 2012, the Coalition government launched the Troubled Families Programme, a three-year initiative that aimed to ‘turn around’ the lives of these ‘troubled’ families (Cameron, Reference Cameron2011). Noted by the Department for Communities and Local Government (DCLG, 2014) troubled families are those who ‘have’ and ‘cause’ problems to those around them, placing high costs on the public sector. For example, it is claimed that troubled families cost the state £9 billion a year, or an average £75,000 per family (DCLG, 2014). As a result, the programme was a policy response designed to address the problems caused by ‘troubled families’, as well as to change the way the state interacted with them (Crossley, Reference Crossley2015). To be eligible for the programme, families needed to include members who met one of the three national criteria for inclusion, namely someone involved in crime and anti-social behaviour; child(ren) who are regularly absent from school and adult(s) receiving out of work benefits (DCLG, 2012). A fourth discretionary criterion allowing local authorities delivering the programme greater flexibility to prioritise families for inclusion who met only two of the three national criteria was also established.
To help local authorities work with the families identified as residing in their area, the DCLG developed a payment-by-results scheme and made available £4,000 for each troubled family with the expectation that the local authority and their partners made up the rest of the investment. A proportion of the £4,000 funding was paid upfront as an ‘attachment fee’ and the rest paid once the local authority had achieved positive outcomes with a family. The success of the programme was to be measured against the eligibility criteria and included fewer school exclusions and improved school attendance over three school terms, 60 per cent less anti-social behaviour interventions and 33 per cent less offending. It also included participation in the Department of Work and Pensions (DWP) welfare to work schemes and/or the end of welfare benefit receipt and the take up of paid work for six months. Finally, success also included reductions in the cost of statutory measures associated with family problems (DCLG, 2012).
Since its launch, numerous concerns have been raised about the programme's role and purpose, as well as its claim of success (Crossley, Reference Crossley2015). For example, the discretionary criteria for inclusion varied between each authority delivering the programme, making comparisons between projects and across the programme problematic (Davies, Reference Davies2015). Hayden and Jenkins (Reference Hayden and Jenkins2014) have critiqued the use of ‘evidence’ in the government's justification for the programme, echoing Gregg's (Reference Gregg2015) claims that social policy development is increasingly being led by ‘policy-based evidence’. Central to such criticism of the TFP are three key issues. The first is that the original figures used to identify the total number of troubled families are controversial and contested. As Levitas (Reference Levitas2012: 4) argues ‘if we interrogate the research behind the imputed existence of 120,000 troubled families, the figure turns out to be a factoid – something that takes the form of a fact, but is not’. This is because the original research on which the figure is based as well as the secondary analysis of the data (Social Exclusion Task Force, 2007) were in no way intended to support the development of such a programme, is outdated and has a number of sampling issues associated with it (Levitas, Reference Levitas2014).
The second is that because of the nature of the payment by results scheme, local authorities have been engaged in extensive data matching exercises to claim success and thus secure monies for families that have had little or no involvement with the TFP. Crossley (Reference Crossley2015: 6) notes that this data matching process ‘involves using available crime and community safety, education and employment data to claim success for families who may have been eligible for the TFP at some stage, but who ‘turned themselves around’, without the support of a key worker associated with the TFP’. Moreover, government support for the development of local discretionary criteria for inclusion of families within the TFP suggests at least tacit approval for this process. The third criticism is that there is a near perfect symmetry between the number of families identified nationally for each local authority, the number of families subsequently found and worked with by these authorities and the number of success claims made by them. As Crossley (Reference Crossley2015) notes, in a period of increasing public funding cuts the claimed 99 per cent success rate makes the TFP an apparently perfect social policy.
Further criticisms have also suggested that such governmental responses to antisocial behaviour (ASB) and family vulnerability are not based on cumulative insight or increased knowledge (Welshman, Reference Welshman2012), but instead are underpinned by assumptions regarding the nature of the problem and the construction of targeted subjects (Ball et al., Reference Ball, Batty and Flint2016). Therefore, another criticism of the programme is that ‘troubled families’ conflates families experiencing multiple problems with troublesome families (Levitas, Reference Levitas2012), thereby implying these families are dysfunctional and anti-social, rather than disadvantaged, excluded and vulnerable (Bond-Taylor, Reference Bond-Taylor2015). As a result, critics have argued that the simplistic criteria used to identify families leads to stigmatisation and the negative connotations associated with being labelled as a troubled family.
Despite these critiques there also exists a body of research, which has identified some common positive aspects to FIPs of which troubled families is the latest incarnation (Batty, Reference Batty2013; Ball et al., Reference Ball, Batty and Flint2016; Boddy et al., Reference Boddy, Statham, Warwick, Hollingworth and Spencer2016). For example, according to Parr (Reference Parr2016), the use of intensive, one-to-one key workers is considered effective for individuals and families who have multiple and complex needs. Research has suggested that part of this effectiveness is because these key workers manage small caseloads that allow for greater time and flexibility to engage and build relationships with families, often in contrast to other services (Sen, Reference Sen2016). Relationship-based practice acknowledges the importance of getting to know people and building up understanding and trust as a precursor to intervention. It has become recognised as fundamental in all forms of family work, for example post-Munro children and families’ social work (e.g. Ruch and Ward, Reference Ruch and Ward2010).
However, despite understanding the need for both time and relationship building in statutory social work, this is often supplanted by increasing bureaucratic exercises and expanding caseloads (Broadhurst et al., Reference Broadhurst, Wastell, White, Hall, Peckover, Thompson and Davey2010). Further, research into social work practice has found that real problems can emerge when social workers do not have enough time to flexibly engage and build rapport and relationships with families due to pressure from high workloads and micro management (Ferguson, Reference Ferguson2014). Research has also found that because of such time restrictions, conflicts are often created with families when scheduling meetings, which can deter families from fully engaging with interventions (Spoth and Redmond, Reference Spoth and Redmond2000).
In summarising the strengths and weaknesses of FIPs, Crossley (Reference Crossley2015: 5) argues that ‘the best that can be said of the family intervention approach, is that it appears to work for some families in some areas of their lives at least for the time that they are supported by a key worker’. Despite this, the current government has deemed the TFP to be overwhelmingly successful, both in terms of helping families and saving money, and announced that the programme would be extended into the current Parliament. To assist with this expansion, the government made a further £200 million funding available for 2015–16 to extend the TFP to reach an additional 400,000 families (Davies, Reference Davies2015). While such investment might at first appear welcome, closer inspection raises some important questions. Firstly, the number of families to be worked with has dramatically increased and secondly the actual financial support per family has been halved, from approximately £4000, to £2000 (DCLG, 2014). Given that some of the most significant advantages of the TFP model appear attributable to a format that depends upon small caseloads and enhanced relationship building, the government must acknowledge that an expansion of the programme into something more large scale risks ameliorating some of the programmes existing benefits. Phase 2 of the TFP may thus inadvertently become self-defeating. Alternatively, it may simply lead to greater data matching as local authorities continue to claim success for families who are not part of the programme to enable them to use central funding to make up for the shortfall created by government spending cuts to other public services.
More recently, an independent national evaluation of the TFP (Bewley et al., Reference Bewley, George and Rienzo2016: 18) reported that overall participation in the programme had no significant or systemic impact. The current article contributes to this debate, by considering evidence about the role of time and caseloads from one particular recent English study of an FIP delivered as part of the TFP. It will then discuss the implications of this evidence in terms of effective practice, success and programme expansion. Before addressing these issues, however, it will look briefly at the research design deployed here.
Methods
This article reports on data collected as part of a wider research study, which evaluated an FIP developed to deliver the TFP in a Unitary Authority in England. Out of over 1,000 families identified as suitable for inclusion within the local project, 122 were being worked with intensively during the time of our evaluation in an attempt to address their needs within the twelve-month period specified by the DCLG (2012).
The evaluation adopted a qualitative approach, employing semi-structured interviews with troubled family workers, programme managers, secondees from other professions and the families on the programme (Wills et al., Reference Wills, Whittaker, Rickard and Felix2016). This approach was used to enable participants to fully explore issues relevant to their specific experience of the project (Bryman, Reference Bryman2012).
Participants
The TFP within the Unitary Authority had a dedicated family intervention team consisting of key workers, secondees from other agencies, supervisors and a team manager. This team was in turn accountable to the TFP co-ordinator at the city council. Overall, fourteen practitioner interviews were held with staff on the programme. These included key workers, managers and supervisors, as well as secondees to the programme comprising a youth worker, a domestic violence worker, a Police Community Support Officer (PCSO), a parenting coordinator and a professional from the Department for Work and Pensions (DWP).
In terms of family interviews, the aim was to speak with families who had been working with one of the key workers who had also been interviewed to compare their different experiences. In addition, as the study was not longitudinal the research team sampled families at different stages within the intervention process to capture the full scope of family involvement with the programme. In total six families were interviewed, sometimes multiple members of the same family were present for an interview, other times family members were interviewed separately. This was a matter of choice for the family but also informed by the researchers’ ethical judgements and overview of the family provided by their key worker. The process of engaging the families in the research is discussed in the ethics section.
Procedure
Each staff interview lasted about sixty minutes. Consent forms and information sheets on the research were given to the participants before the interview. Care and flexibility was required when making contact with families for interview to overcome a range of issues which previous research has described when seeking to engage with hard-to-reach participants (Wills et al., Reference Wills, Whittaker, Rickard and Felix2016). The interviews lasted between thirty and ninety minutes and reflected a conversational style to help participants feel better able to engage; all of the interviews were completed in the family homes.
Ethics
There are clearly a number of ethical and data protection issues in conducting research of this kind.Footnote 1 Because a major part of the research project involved interviewing practitioners and families who are part of the TFP, issues of confidentiality, anonymity, data protection and participant care drove our approach. For example, before interviewing families their suitability for involvement, in terms of minimising any physical, emotional or psychological harm from participation, was discussed with project staff (Bunting et al., Reference Bunting, Webb and Shannon2015). Those families whose suitability researchers and key workers agreed upon were then approached by their key worker to discuss whether they would be willing to participate. The key worker and researchers would then arrange a joint meeting with the family to introduce the researcher and the research (Wills et al., Reference Wills, Whittaker, Rickard and Felix2016). At this meeting, the research was discussed with the family to allow them to ask any questions so they could make an informed, unpressured choice as to whether they would like to participate or not. If they did, the researchers would then arrange directly with the family a time and place to carry out the interview. Participants were given detailed explanation of ethical procedures, including confidentiality and anonymity. However, it was made clear to participants that this did not extend to any issues of concern that might arise about risk of harm to themselves or others. Families were notified at the start of the process that should such issues arise, they would need to be reported to the relevant support agency as is standard when conducting such interviews (Mishna et al., Reference Mishna, Antle and Regehr2004).
Reflection
In terms of limitations of the method adopted, the sampling of families and the implications this has for the findings needs acknowledgement. While recognised practice (Bond-Taylor and Somerville, Reference Bond-Taylor and Somerville2013) – using project key workers as gatekeepers to help identify and access families – may have led to sampling bias, with only those families who were successfully engaging with the project selected for interview. However, the importance of adopting an ethics led approach was central to our sampling strategy, and while more diversity in opinion may have been obtained with a more random sampling method, care of potentially vulnerable subjects might have been sacrificed.
Analysis
A thematic analytical approach was adopted for the interview data in this article (Braun and Clarke, Reference Braun and Clarke2006). All recordings from the interviews conducted were transcribed in full into word documents. All material was then read by the researchers and discussions held around issues raised by the data to create themes. The data within each theme were then reread with sub-themes emerging, which form the analysis presented here.
Results
As identified earlier, previous discussions of FIPs, including the TFP, have tended to highlight small caseloads, worker flexibility and relationship building as their strengths, and this was also found in the current analysis. Additionally, analysis also found that helping service users find a voice to fight their own battles and get a hearing with other professionals was important, as was the capacity to help service users break down loneliness and isolation. Time seems fundamental to nearly all of these aspects – the commodity lacking in welfare services and directly linked to funding and organisation. The article now considers the particular areas of strength identified in more depth.
Time and flexibility
This evaluation established that a valued aspect of the key worker approach to working with families was their ability to be flexible and spend time with the family. This was highlighted as important in beginning, effecting and even ending the work. For example, within the project evaluated, time was created by management for key workers to enable them to understand and make plans to address particular families’ specific needs. This tailoring of the service to the family is in contrast to other services that have a more rigid format and schedule which the family are required to meet in order to access them (Spoth and Redmond, Reference Spoth and Redmond2000). This can be problematic where families are chaotic and facing multiple issues and crises. Additionally, lack of confidence, poor mental health and the sheer number of difficulties being faced made it hard for families to tackle often very small and straightforward problems. Many situations were complex and chronic, as one family member noted:
I've got five children altogether … [child 1] who has just been diagnosed with autism, [child 2] was shortly diagnosed with autism, ADHD and ODD . . . we've had a whole lot going on with our family in the last two months … we lost my Dad at the beginning of February, my Mum's sectioned under the Mental Health Act. (F3)
The initial engagement with the family appeared successful because of time and the ethos of the project. Key workers used persistence to gain access to families, often going to the family home every day and phoning and writing. This was noted to have positive results by key workers:
I think having the time, the time to be able to go to families as many times as they need you to is helpful, I think having the flexibility to use you know resources and work creatively. (KW3)
According to key workers, a crucial aspect to their work was having a small caseload, which meant each family could be worked with according to their needs as opposed to the operational constraints of the service. This was something that key workers also believed benefitted other agencies who were working with the family:
I think they find it quite handy to work with us because we can, we have the capacity to go round more, because of our smaller cases … but then equally we need them to do the, you know, to kind of lay down the laws and the legal side of things … I guess we complement each other. (KW1)
Families also valued the capacity of the worker to go at the pace they needed, and to offer the support in a way that they could manage:
if I just said to (KW), you know what I really need is to get out and have a coffee, then she would take me out for a coffee … and gradually in the last year I've felt able just bit by bit to go to the shop myself … just to go out in the garden … I didn't think I was ever going to get out the flat again. (F6)
Relationship based practice
Closely related to the issue of time is that of building positive relationships between workers and service users. Being able to spend time with families due to low caseloads enabled the key workers to develop relationships with the families. Having this time allowed families to get to know and trust the key worker, and gave them the confidence to be able to share information that they may not ordinarily have shared with other services. As one parent explained:
If they don't have the time to gain that trust, it's almost as though they come and they want this, this, this and this, and this criteria to meet … and (the child) might need to gain trust because he has none. (F3)
Within the twelve-month period that key workers had to work with families, it was their flexibility to use time to develop relationships of trust that helped them support families and break down isolation. As a family member noted:
I think the last few months I would probably have had a breakdown if I didn't have [KW], one sane person there in the corner saying, you know, saying perhaps you should try this. (F3)
Families reported the biggest impact often arose from the smallest actions. Being valued helped to change their perceptions of themselves:
(KW) used to ring me up going ‘well done for going to school today’, or ‘looking forward to the next day’, (KW) would text me in the morning, ‘hope you have a good day in school’ . . . like stuff like that, just encourage me, and that used to make me think yeah I am doing it. (F4)
Families also discussed how it was not just the amount of time that the key worker spent with them that had a positive impact, but also what the key worker did during this time:
It's only a two-minute phone call to see how I am, to see that I've got all the paperwork ready, and it's those things you know that does matter. It doesn't matter how long they spend with us or anything like that, it just matters what they do within those minutes. (F1)
Having the time to work with and get to know the families, to understand the issues they were facing, also allowed the key workers to recognise where a family may need more time or extra support from the programme and other agencies. A key worker gave the following example:
Social workers were saying that they didn't feel the need to keep the case open, so I would have been the only person kind of there for the family, and (the agency) extended their time for a month at my request and the social workers stayed involved at my request. So I think it's just good to know that we have the ability to ask people and kind of just be influenced maybe to say look we know that there are a lot of needs for this family and we know that you need to be involved and this is why. (KW6)
Confidence and voice
Overlapping with many of the sentiments expressed above, and also connected no doubt to time, flexibility and a service user ethos, is the quite crucial issue of encouraging service users to develop confidence to address systems in which they feel unheard or disrespected. Empowerment is an often used concept in social work, with contested definitions and practices (Bond-Taylor, Reference Bond-Taylor2015), but assisting people to build their confidence so that they can ‘fight their own battles’ is at the heart of this. This may start with small steps between the worker and service user (Houston, Reference Houston2016). Spending time with a family appeared to make them feel valued and heard:
it gradually took time I think … so that's made me think like she has faith in me, like so I was more myself. I was thinking, someone believes in me and like I'm glad she believes in me and I didn't use to believe in myself until she told me I can do it. (F3)
By contrast, families identified how they felt that they were not being listened to by other services:
if you as a mum are going round to all these services and nobody's hearing what you're saying, it's frustrating and then you end up giving up, which is what happened with my eldest son. (F1)
Signposting families to get support and providing direct guidance to them to help them achieve the outcomes they needed was linked to feelings of having their voices heard:
I had problems with . . . stuff in the flat, general repairs, (KW) would ring up and get them sorted . . . if I couldn't make an appointment because I've got a phobia of phones . . . but I'm getting better with that, you know I'm able to ring up my psychiatrist … and say if I'm not going to be able to make it, so that's good I think because I've not been able to do that for quite some time. (F2)
Families in this study reported becoming better able to directly address areas where they had felt ‘unheard’, such as schools, health workers, social services etc. A family member commented:
They helped mum a lot … her confidence … since they've been on the scene even with the schools and everything like that, she's getting confidence. (F6)
Issues of time, then, and its concomitant flexible working practices, as well as the approach of the service, seemed to be particularly conducive to providing person-centred, supportive and individually tailored work that could encourage service users in ways that were obviously empowering. In addition, and perhaps a crucial feature of making all this possible, this study found that time and again the families associated the key worker with helping them break down the isolation they experienced. Loneliness and isolation are complex issues, but are beginning to be understood as of fundamental importance to people's well-being and mental health (Frost and McClean, Reference Frost and McClean2014). Certainly, the issues recurred here. For example, a service user commented:
My world was the front door and the back garden … didn't want to go anywhere else … he [key worker] gave me the right encouragement to get through the door. (F3)
This was further elaborated by another woman, for whom lack of confidence and isolation were part and parcel of family issues. Talking about going to a parenting group, she said:
Like the first time [the worker] came with me, and she dropped me in it (laughs). She said ‘right, I'm going now’ (laughs), and I'm like God! And like I'm sat there but as soon as I started speaking to people I loved it. It was like every Friday my time, like he [her partner] had the kids and you just go there, you do what you have to do, you have like a cup of tea when you want, go out for a fag when you want. (F6)
The analysis of the data has highlighted how giving troubled families’ time and attention appears central to success. As one key worker put it:
There is definitely something in the amount of time we can give, feels like I trained to be a social worker and this is what I imagined the social worker's role to be and I know in reality, this is not what social work is like, it's so full of paperwork and statutory stuff, um, they don't have time to do what we do. So I know I kind of feel privileged that I am actually able to do this work and that maybe it's because I think it's the most effective way of helping families. (KW1)
Having the time to do the job and the range of skills and practices it entails appears to make effective interventions possible, and can assist in creating positive change for families. According to another key worker:
It's about changing their circumstances and giving them real opportunities, but also enabling those real opportunities to happen, and to give them the confidence and the self-esteem to make that change takes time and there is no short term fix to this if you really want to look at your complex need families. It does take time and there is no point in saying we are going to eradicate this in a year because you are not going to do that – it takes a more complex structured programme to enable real change to happen. (KW10)
Conclusions
Analysis identified a range of good practices similar to those found in previous FIP research (Parr, Reference Parr2012; Batty, Reference Batty2013; Hayden and Jenkins, Reference Hayden and Jenkins2014), and additionally uncovered how key workers were able to empower and connect with service users. The article also identified the strength of feeling service users expressed in their preference for this form of service delivery over local authority social services. Key workers saw their practice in rather idealised terms as what social work should but cannot currently do (Featherstone et al., Reference Featherstone, White and Morris2014). Both key workers and families discussed how the time and flexibility afforded by the programme enabled relationship-based practice to develop (Parr, Reference Parr2016). This practice, we discovered, also appeared to give families confidence and voice (Houston, Reference Houston2016) and helped to break down barriers of loneliness and isolation (Frost and McClean, Reference Frost and McClean2014).
In terms of limitations, this project has not examined the potentially stigmatising effects that being part of the TFP may have on families (Bond-Taylor, Reference Bond-Taylor2015). It also reported on data that were largely positive, focussing on the successes of the project without presenting opposing opinion. As identified, the sampling method may have influenced this. Therefore, care should be taken in generalising these findings beyond the context within which they were obtained (Starkey, Reference Starkey, Borsay and Shapley2007). However, the factors discussed as positive by families and workers chime well with that which previous qualitative research of FIPs has identified.
Despite such qualitative evidence of the practical benefits created by FIP's, the government's championing of the TFP appears instead to have focused more on quantitative data to argue for its overarching policy success. This is especially problematic in light of the recent independent evaluative report of the TFP (Bewley et al., Reference Bewley, George and Rienzo2016) that found no quantifiable evidence of such positive results. Positive but general government claims, potentially boosted by local authority data matching (Crossley, Reference Crossley2015), have therefore paved the way for the expansion of the TFP into phase two without a clear understanding of what works within the projects set up to deliver it. Bewley et al.’s (Reference Bewley, George and Rienzo2016) identification of ‘some signs of green shoots’ in the TFP, including that the families involved experienced levels of increased confidence and optimism, suggests that a nuanced understanding of what works in delivering FIP's as part of the TFP is vitally important for its future.
Without this understanding, a focus on the quantification of success or failure over qualitative tempering may ultimately prove self-defeating for the TFP. For example, the second phase has seen a near fourfold increase in families to be identified and worked with whilst halving funding available to do so. Given the already short twelve-month period for ‘turning around’ families (DCLG, 2012) increasing caseloads will further reduce the time available for each, which may then ultimately lead to the programme's failure. As phase two of the TFP continues, it is vital for researchers and practitioners to continue to reinforce the significance of having enough flexible time (and hence small caseloads) so that the programme's so-called success does not come at the expense of what actually helps families, as it expands at a rate not commensurate with such practice.
