Whole Family Approach: rapid review of literature

4. Literature review

4.1 How are families defined?

How families are defined has changed over time within the academic literature, with implications for the implementation of a 'whole' family approach. While the family unit was traditionally defined and understood as being heterosexual, living together and determined by biological ties (Williams, 2004), contemporary sociological thought has highlighted an increasing diversification of family compositions and practices. This has included an increase in single parent families, cohabitating parents, same-sex parents, transnational families and changes in the gendered caring norms within families (Walsh and Mason, 2018). Such a conceptual transition has been underpinned by a shift from a definition of the family as being purely biologically determined to also being socially constructed (Morgan cited in Walsh and Mason, 2018). This reflects the ways in which families report experiencing relatedness, which increasingly includes non-biologically related individuals (Smart, 2007) and a focus being placed on the practical contribution made by different individuals to family life and caring responsibilities through the roles performed and tasks undertaken (Walsh and Mason, 2018).

Notably, in spite of this expanding definition of the family, a gendered asymmetry is often highlighted in the literature (Mendoza et al, 2010). Some literature reports that emphasis is routinely placed on the mother-child relationship (Walsh and Mason, 2018) and the effect of a parent's substance use on children has been reported to be considered more severe when it relates to a mother as opposed to a father. This is despite research emphasising the importance of the father-child relationship (Lewis and Lamb, 2007). As a result, mothers who use substances may face additional barriers to accessing therapeutic support due to increased feeling of shame or guilt, fear of having their children removed from their care and feeling they are subject to increased scrutiny and judgement regarding their capacity to care for their children (Thompson, 2022). This can in turn lead to the development a deep-rooted reluctance around support from social services, thus creating a further barrier to treatment and support (Thompson, 2022).

In the context of implementing a Whole Family Approach (WFA) in Scotland, the definition of the family needs to capture the reality and complexity of all family structures in order to successfully achieve its aim of supporting the family as a whole as well as its individual members. An incomplete understanding and definition of the family risks withholding support from individuals affected by a loved one's substance use and failing to involve the right individual's in a person's substance use treatment. Research with people with lived experience has shown that this can result in feelings of distress and defeat as individuals are forced to work with multiple different, connected yet distinct services and with professionals addressing the needs of individuals rather than those of the family as a whole (Webb et al, 2014; Gentile and Clapton, 2021). The existing evidence base acknowledges that these are longstanding issues that affect how service users interact and engage with WFA practices and which therefore need to be addressed by substance use practitioners deploying these interventions.

The evidence base on WFA in the context of substance use emphasises the role of families in treatment. Morris et al (2008) see the family as a basis of support for family members with problem alcohol or drug use. It is acknowledged that substance use affects families in numerous ways by affecting rituals, roles, routines, communication, social life, finances, relationships and interactions (Velleman and Templeton, 2007). Exposure to parental substance use can shape a child in a number of ways. The literature emphasises that children and young people are particularly vulnerable to the effects of substance use in their family and primarily focusses on the negative outcomes. Children and young people may be subject to experiences of violence or neglect, have to adopt responsibilities or parenting roles at an early age and can experience negative emotions such as shame, guilt, fear, anger and embarrassment. Children may also experience neurodevelopmental effects if exposed to substances in utero (Velleman and Templeton, 2007; Mitchell and Burgess, 2009; Templeton, 2012; Gentile and Clapton, 2021; Ordord et al, 2010). This may result in behavioural or emotional difficulties, such as low self-worth, reduced confidence, difficulties in forming relationships with their peers, underachieving in school (due to the impact of correlated poor attendance and lower involvement in extra-curricular activities), social isolation, premature maturity, or early alcohol and drug use (Templeton et al, 2006; Velleman and Templeton, 2007).

Family support has acknowledged the wide range of effects substance use can have on whole family systems, resulting in the adoption of three types of family interventions. These are (1) services for families that focus on entry to treatment for the person using substances, (2) practices that include the direct involvement of family members in the treatment of the person using substances and (3) services that aim to meet the needs of affected family members in their own right (Copello et al., 2005). Despite there being strong evidence that family members are negatively affected by a relative's substance use and it is also true that family involvement reduces distress, enhances coping skills and improves family functioning (Kourgiantakis et al, 2021). Current research has however focussed predominantly on substance use outcomes and includes less information about other outcomes (Copello, 2006; Velleman et al, 2005).

4.2 How is substance use understood?

While the evidence base acknowledges the longstanding effects of substance use on families, it also identifies entrenched narratives in existing family support that place a primary focus on substance use to the possible exclusion of addressing wider needs. For instance, research suggests that social services can focus their attention on a particular issue without taking into account the broader familial context, or consider an issue as affecting an individual rather than an entire family unit (Gentile and Clapton, 2021).

A number of publications challenge the way in which substance use is understood and approached within family inclusive practices. Rethinking the drivers and experience of substance use is key in informing the formation of a WFA. This is explored in relation to challenging existing institutional cultures that take a risk averse approach to assessing substance use, with family support for the most part being seen as a reactive, deficit model, where parents who use substances are defined in negative terms (e.g. the risk and harms they may pose to their children) and where positive aspects are disregarded or undervalued (Tew et al, 2016; Morris et al, 2017).

Despite a public health approach to substance use adopted in Scotland – in contrast to the criminal justice model that underscores UK Government policy around substance use – research in a Scottish context often makes reference to UK policy, suggesting that this may still affect overall attitudes. The evidence suggests that this may present people who use substances as failing to exercise good judgement in taking up opportunities to become hard working individuals and families (Featherstone et al, 2016; Tew et al, 2016; Morris et al, 2017). Moreover, this places responsibility at an individual level, frames substance use as inherently aberrant and reflecting of an individual's general failing, and disempowers individuals by not taking into account any positive contributions. It is argued extensively that this deficit model of substance use diverts attention away from systemic drivers of harm (Gupta et al, 2014; Featherstone et al, 2017; Flacks, 2019) and, as noted above, is intrinsically gendered (Thompson, 2022).

There has been a shift towards acknowledging the wider context in which substance use occurs within the evidence base. Increasingly the argument is being made that the wider context in which substance use occurs and of which it can be a symptom needs to be considered and taken into account when providing support to individuals and families affected by substance use (Gentile and Clapton, 2021). Kroll and Taylor (2009) see drug use as symptomatic of a complex range of longstanding psychosocial factors as well as being the cause of additional difficulties, while Forrester et al (2013) acknowledge that a range of issues, including poverty and deprivation, can lead to poor childhood outcomes. Further, Velleman et al (2005) argue that family influence does not exist in a vacuum. There are also other important determinants in substance use, including: intra-personal factors, peer influence and wider community and environmental factors such as media influence, advertising, and environmental deprivation that must be taken into account in understanding the causes of substance use and in corresponding approaches to prevention and intervention practices (Velleman et al, 2005). Family support is therefore seen to sit within a changing policy landscape that has moved from a deficit framework of parenting to one that acknowledges and seeks to address these wider drivers of substance use, seeing substance use as a symptom of larger systemic issues.

4.3 How is Whole Family Approach defined?

Attempts to reframe understandings of the family and of substance use more generally – moving away from deficit narratives of substance use as inherently aberrant to considering its wider context and drivers – have acted as a foundation for the development of the core principles of a whole family approach. However, family support as a whole (within which WFA sits) is largely poorly theorised and articulated, with few studies including a clear and explicit definition (Devaney and Dolan, 2017; Leonard et al, 2018). It has been described as a 'slippery concept' (Frost et al, 2003) and there is a lack of consensus in the literature in terms of what it entails, when it is applicable, its value as a service or the practicalities of its implementation (Devaney and Dolan, 2017).

A lack of definitional clarity was evident in the initial scope of the literature, revealing an array of differing terminology that broadly spoke to practices adopted in family support. However, an assessment of these approaches revealed broad conceptual crossovers in terminology, with no clear differences in ethos or application. From the outset, it was clear that there was limited conceptual clarity or distinction as to what constitutes whole family approach or family inclusive practice, or how they differ from the likes of family-focussed practice or family-based interventions. As a result, the evidence base works around the iteration of a series of core principles that cut across a range of family support interventions and broadly define these as a "continuum of advice, support and specialist help starting in the community and signposting the family towards early, less traumatic interventions" (Audit Commission cited in Devaney and Dolan, 2017: p.11). Family support seeks to give children and young people the opportunity to reach their full potential by helping and empowering families and strengthening communities (Devaney and Dolan, 2017). It is underpinned by the principles of prevention, strengths-based, developing resilience, realising rights; and rooted within an ethos of collaboration between professionals, people who use substances and their family members (Pinkerton et al, 2004; Devaney and Dolan, 2017).

Taking into account he above, and for the sake of clarity, this report uses 'whole family approach' as an umbrella term for the range of terminology used in the literature (e.g. family-focussed work, family inclusive practice) and therefore applies this term when discussing the evidence, regardless of what terminology was used in the publications. Whole family approach, as described in RRR, is taken to mean an approach to service provision that is co-ordinated, holistic and offers support to families as a collective unit, but also acknowledges that individual members of the family are affected by substance use as well as a range of other factors, as discussed in this report.

This definition acknowledges that individual family members are in need of support in their own right (Gentile and Clapton, 2021). It is generally agreed upon that WFAs should be underpinned by a commitment to giving families the tools to work together to identify their needs and generate solutions and goals (Devaney and Dolan, 2017; Scottish Government, 2021). All members of a given family (however that family defines itself) should be consulted and included at all stages of service provision and evaluation, thereby ensuring that services capture diversity across race, gender and sexuality in providing positive and consistent support for all families in a responsive and safe way. This should be balanced with being mindful of the specific needs of individuals within the family, where necessary intervening to safeguard individuals, and taking into account the strengths and fragilities of relationships (Scottish Government, 2021). A WFA should involve multi-agency collaboration that acknowledges and challenges systemic barriers to participation, the stigma experienced by people who use substances and their families, and the gendered effect on parental rights (Scottish Government, 2021).

Within the context of substance use a WFA is linked with approaches to treatment and support that acknowledge the centrality of structural factors in a way that is often neglected in the individualised assessments and interventions that typically characterise family support (Gentile and Clapton, 2021). As Morris et al state, a whole family approach ought to 'understand [that] the parents and children's difficulties are more often […] a function of exclusion rather than a cause' (2008: p.83).

When the wider context in which substance use and its associated harms occurs is considered, responsibility can no longer be placed at an individual level. Central to this is emphasising the state's responsibility (as opposed to the parents) assuming experiences of harm are the product of wider social determinants, such as poverty (Featherstone, 2016).

Underpinning a WFA is a broader commitment to family resilience e.g. the ability to adapt or to overcome adversity (Gentile and Clapton, 2021) as a means of safeguarding children's wellbeing. Family resilience is defined as a family's ability to 'maintain its established patterns of functioning after being challenged and confronted by risk factors', and reflects 'the family's ability to recover quickly from trauma or a stressful event causing or requiring changes in organisation of the family' (Kalil, 2003: p.11).

A WFA aims to build a family's resilience by giving them the space to identify their own problems and to coproduce their own solutions, implementing restorative practices, and supporting them to find constructive solutions (Morris et al, 2017). This is largely underpinned by the core principle that family support should be strengths-based, i.e. drawing on the existing strengths of families, challenging cultures of blame and encouraging parental empowerment (Thompson, 2022).

This is paired with demands for a WFA to be trauma-informed. WFAs are seen to offer innovative, proactive and preventative models in contrast to traditional medicalised narratives of diagnosis and treatment. This is an approach that acknowledges and responds to trauma (identifying signs and symptoms) and works collaboratively to ensure that services do not re-traumatise individuals. Instead they facilitate access to counselling or group support, and provide safe environments based on trust, collaboration, choice and empowerment. A trauma-informed approach also challenges asymmetrical power relations between service users and workers by prioritising strengths-based practice and encourages a whole organisational shift in ideology (McCarthy et al, 2020). WFA interventions should be person-centred and recognise that pathways to recovery are complex and different for everyone, and may be accompanied by setbacks and relapses (McCarthy et al, 2020).

Finally, a WFA is underpinned by a commitment to the realisation of human rights. Specifically, a WFA seeks to affirm the dignity and equal rights of all members of the family (Featherstone et al, 2016). Children's rights, as per the United Nations Convention on the Rights of the Child (UNCRC), are entangled and inseparable from those of their parents, family and community as the latter are considered essential to the nurturance, fulfilment and safekeeping of children.

4.4 Examples of good practice

There is an extensive body of literature assessing the effectiveness of family based interventions in the context of substance use. They are predominantly academic and grey literature reviews of pilot projects and have focussed on describing the practicalities of family support interventions in reducing drug-related harms and exploring their outcomes across a range of substance use, social, emotional and behavioural domains. This section explores these studies in detail, focussing on examples of best or good practice with relevance to the Scottish context.

It is important to note that the majority of the literature on best practice consists of evaluations where the effectiveness of the service as a whole is assessed, without necessarily differentiating between the multiple different techniques and strategies used within these interventions to engage with different members of the family. This means that while there is a clear indication that WFA interventions work, it is not necessarily possible to attribute positive outcomes to specific elements of the interventions. Moreover, variations in the methodology employed by the different research studies (e.g sample size, use of qualitative and quantitative empirical material, location of each study) makes direct comparison difficult.

4.4.1 Involving families in design and delivery of services

The literature suggests that support services designed and delivered in consultation with families resulted in goal specific, measurable and realistic interventions. The evidence specifically emphasises the importance and effectiveness of not judging or giving up on families, treating them differently to how they are accustomed to being treated, identifying what motivates them and making use of existing strengths at an individual and family level to collaboratively develop treatment or support plans tailored to their needs (Mitchel and Burgess 2009; McCarthy et al, 2020; Circle, 2020^[4]; Thompson, 2022). Successful interventions often involved an initial period of engagement and assessment with the entire family to identify their needs at an individual and family level. Key to this approach is engaging with families both as collective units with shared needs and as individuals within that unit with their own specific needs (Burgess, 2011). Where family members included children, focus was often placed on helping parents understand their children's needs (Mitchel and Burgess, 2009; Copello et al, 2005; Gentile and Clapton, 2021). Children should be provided with opportunities to express their feelings and experiences about the family situation and may benefit from being facilitated to contribute in discussion with their parents and adult members of the family (Mitchel and Burgess, 2009).

The evidence links involving families in the design and delivery of WFA interventions with the adoption of therapeutic approaches and relationship-based practice. Evaluation research suggests that these contribute to positive working relationships between professionals and service users, with the later reporting feeling less judged and perceiving professional staff as more honest, helpful and attentive (Alderson et al, 2022). This is supported by other research that found that the use of restorative approaches in family services promotes whole family, relationship and strengths-based services. A restorative approach is underpinned by the ethos that repairing or resolving harm is best achieved by building on or strengthening relationships rather than penalising those involved (Williams, 2019). This commitment to resolving harm is enacted via a commitment to collaboration, voluntary participation, respect, trust, non-discrimination, safety and honesty.

Successful interventions that worked collaboratively with families and the person using substances included network therapy, behavioural family counselling, and motivational enhancement therapy, solution-focussed therapy and motivational interviewing among others (Williams, 2019; Kourgiantajis et al, 2021). The former are types of relapse prevention interventions utilising the support network of the person using substances (Galanter et al, 2002; Glazer et al, 2003; Galanter et al, 2004; Copello et al, 2006), while the other interventions seek to develop resiliency and strengthen support networks around the person using substances. Forrester et al (2008) argue that techniques such as motivational interviewing enable professionals and families to address resistance, behaviour change and confrontation – common issues that need to be challenged in the pursuit of good practice.

The importance of involving families in the design and delivery process is demonstrated in Gentile and Clapton's work (2021), which found that there was a clear disconnect between how families and referrers assessed and prioritised needs. Specifically, referrers considered supporting children's emotional and behavioural needs to be the top priority, while families tended to identify unmet material needs (e.g. poverty, housing, benefits and debt) as their primary concern (Gentile and Clapton, 2021). Central to this discrepancy may be that referrers were interpreting substance use as the cause of family issues as opposed to a 'symptom' of wider driving factors (e.g. poverty). This wider context in which substance use occurs also needs to be considered in order to meet the needs of the family.

4.4.2 Long term and time-sensitive

The importance of both appropriately timed and long term interventions was a key theme in the literature with regards to best practice. A practical issue identified in the evidence base was that all interventions operated on a relatively short term basis due to conditions surrounding their funding. The majority of projects lasted between 4 weeks to three months, with the longest running for just over a year (Murphy et al, 2010). This variability in the duration of interventions was raised throughout the evidence base as a matter of concern, and the effectiveness of WFAs in producing positive outcomes was largely ascribed to the provision of long-term and time sensitive interventions.

Service users were commonly cited as reporting that they wished for projects to run longer or that they felt that they needed longer term support, with Copello et al (2006) finding that families are often "left hanging" when they reach the threshold for participation. Furthermore, Templeton's (2012) exploration of the MPACT project found that families wanted the project to continue after the 8 week period as they felt that they needed ongoing support or would like to include other members of the family. Participants also felt that the sessions themselves were rushed. The short term nature of certain interventions made participants feel like they were unable to maintain changes in the long term and that children, specifically, were left without support afterwards (Templeton, 2012). This concern is echoed by a number of researchers who argue that a short term "fix" is insufficient and that services must to be designed to allow families to return for a "top-up" of support (Burgess, 2011; Tunstill and Blewett, 2015). A longer timeframe for "bedding-in" is also highlighted as crucial for effective inter-agency collaboration (Alderson et al, 2022).

In addition, in order to be effective, a WFA must be taken from the outset rather than as a last resort for families in crisis and should not distinguish between families with children and those without in terms of importance (Tew et al, 2016). While in ethos WFA seeks to offer family support as a preventative measure (see Kroll and Taylor, 2009), it was not uncommon throughout the evidence base for support to be provided only to families at crisis point (Nagle and Watson, 2008; Mitchell and Burgess, 2011; Thompson, 2022; Alderson et al, 2022).

This was of major concern to Burgess (2011) in the Glasgow Bridges pilot project where the targeting of families at the early intervention stage became rather difficult due to a low rate of referral to service, resulting in the need to expand inclusion criteria to include children under 14 years of age from surrounding geographical areas. Furthermore, the decision was made to extend invitation to "kinship families" for a six month period, suggesting that the initial project did not include different "types" of family composition. However, those who required early intervention often had had support needs due to deep-seated or hidden difficulties (Burgess, 2011). While most of the research consulted emphasises the importance of preventative practice, it is clear that early intervention can be a challenge for services that are dependent on short-term funding.

Linked to long term interventions is the emphasis placed on the establishment of positive therapeutic relationships, with significant contact time between service users and professionals (Mitchell and Burgess, 2009). Research emphasised that it takes time to build trusting and supportive therapeutic relationships (Burgess, 2011) but that successfully doing so is associated with an increased in the likelihood of positive outcomes, including a reduction in substance use (Raistrick et al, 2006). Forrester et al (2013) found there to be a marked reduction in the number of children entering care, a reduction in substance use, improved parental and broader family wellbeing and family cohesion. While the length and intensity of the programme and relationships was not directly tied to these outcomes, they were important aspects of a whole service that worked alongside other elements to produce positive outcomes.

The research suggests that a number of aspects contribute to building a therapeutic relationship, including adopting a nurturing approach with service users; adopting a flexible approach that fits with the fluctuations of family life; and acknowledging that some issues may take longer to address and that not all objectives will be short-term (Burgess, 2011). Research emphasised the need to adopt flexible approaches, tailored to the individual circumstances of service users. Some interventions built in flexibility in the way staff organised their availability to service users, for example by maintaining an open door policy to service users during office hours or by offering phone calls and text messages as well as in-person meetings to facilitate access to professional support in a timely manner in-person contact time (Mitchell and Burgess, 2009).

4.4.3 Taking a holistic approach

The wider context of problem substance use (e.g. structural or systemic inequalities and ecological drivers) that create barriers to accessing and engaging with support services should be acknowledged, responded to, and challenged for effective whole family working. However, while academics and policy makers in the field acknowledge the contextual drivers of substance use and the evidence base shows the importance of adopting ecological approaches to service provision that address systemic inequalities of which substance use is a symptom (Gentile and Clapton, 2021), few projects acknowledge that such inequalities may act as barriers in engaging with family support services.

An exception is found in a Welsh mixed methods study of a service that worked with children at risk of entering care due to a parent's substance use (O'Connor et al, 2014). The research involved 27 families (84 children) and investigated the effect of short term intervention services. All families were referred due to child welfare concerns having reached statutory child protection thresholds. The project replicated a model of American "homebuilders" interventions in which practitioners were available for 24 hours a day and provided domiciliary care to families over 4-6 weeks. Therapeutic techniques used included motivational interviewing and solution focused brief therapy (O'Connor et al, 2014). Key to the project was challenging the binaries at play in narratives of family support, which lead to focus being predominantly placed on child protection or substance use related needs, thus losing sight of the intersectional nature of needs (O'Connor et al, 2014). The project found that narratives around substance use and societal perceptions of parental neglect could not be disconnected from the material factors noted in participant interviews (e.g. poverty, poor housing, domestic violence) and emphasised the importance of integrated models that recognise internal and external determinants. These were key to either promoting or hindering family engagement, developing nuanced understandings around the drivers and causes of substance use, and developing a specific toolkit of approaches (e.g. motivational interviewing or direct work with children) to strengthen family interventions (O'Connor et al, 2014).

Significant attention within the evidence base has been given to the gendered nature of parenting and the ways in which this becomes exacerbated in the context of substance use. For example, some research indicates that women may experience higher levels of fear around the loss of a child into the care system and their substance use may be more likely to be seen to negatively affect their parenting than their male counterparts (Thompson, 2022). This is seen in Thompson's (2022) qualitative mixed method evaluation of a WFA project in the North West of England that sought to offer support to parents, children and families affected by alcohol and substance use by exploring the way mother's make sense of their substance use and their experiences with various interventions that were perceived to either have helped or hindered their individual journeys to recovery. This identified the importance of family-based interventions that work with individuals in non-judgemental and empathetic ways and consider the needs of mothers as well as those of the family as a whole.

The need to acknowledge the systemic drivers of substance use and challenge individualised narratives around substance use is an overarching theme in the literature and key to doing so is the provision of holistic support for non-substance use specific needs, sometimes described as "practical support".

An example of practical support provision is found in the Glasgow Bridges project, which delivered family-focused interventions to improve childhood resilience to adversity and, in turn, improve educational, health and wellbeing outcomes (Burgess, 2011). The project worked with 30 families (53 children up to the age of 14 and primarily their mothers/female carers) between 2009 and 2011. Children were referred for issues related to poor social skills, poor education outcomes, concerns around emotional attachment, and physical ill-health. Parents were primarily referred to the project for their substance use and a need for help with establishing boundaries and household routines. Referrals also noted low self-esteem, fraught family relationships, poor mental health and experiences of domestic violence as common issues for the families as a whole.

The project adopted an ecological approach, acknowledging the intersectional drivers of the situations the families found themselves in such as poverty, poor and insecure housing, the struggle to provide emotional, social and educational support for children and substance use (Burgess, 2011). It prioritised the promotion of resilience, developing the child's holistic wellbeing and emotional development, as well as educational attainment. In order to do so, the project offered a 'menu' of interventions for service staff to pick from to use with families. These included early morning support to establish routines to get children ready for school on time; individual and group parenting work; joint adult and child play sessions; educational support through individual and group work (e.g. homework support for children and their parents); and organised family outings. The findings suggested that interventions were most effective when approached creatively, such as by offering practical tasks like baking, crafts and playing board games to teach literacy, numeracy, sharing and communication skills (Burgess, 2011).

By the end of the project, it found that most families had made some progress in most but not all areas (Burgess, 2011). Over a third (37%) of families showed improvement across the main outcomes measured. These included experiences at school; health and wellbeing; and parenting awareness, skills and confidence (Burgess, 2011). These outcomes were found to be directly connected to the interventions as a whole but the evaluation methodology does not allow for an assessment to be made concerning what specific practices within interventions were successful in producing positive outcomes. For example, school attendance rates were improved through early morning 'ready for school' support, while children's ability to manage the school environment, concentrate in class and co-operate with peers was supported by individual and group work. Improved child-specific outcomes included increased signs of confidence, greater ability to express feelings, increased participation in social activities and support networks, reduction in risk associated with having a substance using parent and in relation to their safety at home. Parent-specific outcomes included increased awareness of their child's needs (including diet and exercise needs), enhanced ability to communicate with their children and to manage behaviour in positive ways, demonstrated greater interest and ability to play creatively with children, school reports of increased involvement in their child's education, and increased confidence in their ability to make positive choices about their own and their family's future. Finally, there was some degree of reduction in substance use or significant reduction/cessation, which parents attributed to the support and encouragement they received from staff and other service users.

Another example of good practice with regards to offering holistic support is identified in a report documenting the result of action research designed to both understand and evaluate the experiences of those receiving whole family preventative support from Circle (Gentile and Clapton, 2021). Circle is a Scottish third sector family welfare agency working with over four hundred families and children, providing services to families on the cusp of formal child protection in East Lothian.^[5] Based on the participation of 35 families over a period of 9 months (although working with the small sample size of 10 semi-structured interviews with parents and only 6 questionnaires completed by children and young people), it was found that offering practical support, focussed on the whole family and which primarily involved working with outreach support workers, helped families identify and address the challenges identified at the start of their involvement with the project. This included providing emotional support and showing commitment to the family by 'going the extra mile' through active listening, practical and material support (e.g. buying bedding, furniture), accompanying families on day-to-day activities like shopping or appointments, referring families on to additional support services where needed (e.g. general practitioners, nurseries, schools, welfare benefits, housing and childcare) (Gentile and Clapton, 2021: p.56). In evaluating Circle's whole family services, Gentile and Clapton (2021) found that after the intervention 43% of families had resolved all of the issues identified during the referral, 40% had resolved most or part of these, and 17% had resolved at least one.

4.4.4 Outreach workers

Central to the offering of practical support for families in effective whole family services was the provision of outreach support. The effectiveness of having outreach workers working closely with families was routinely noted by Mitchell and Burgess (2009) in their review of WFA interventions. Outreach workers may contribute to reducing barriers to access and promoting engagement by facilitating access to other services, for example by helping families navigate the system or by attending appointments to resolve issues affecting the family generally and beyond substance use (Gentile and Clapton, 2021). Families report that having someone making the complex procedures of formal child protection understandable to them and giving them the confidence and skills to advocate for themselves is important (Gentile and Clapton, 2021).

Most WFA activities were undertaken within the community and, more specifically, in the home of service users as engaging with families 'on their ground' was considered less intimidating. While interventions delivered in service users' home are not without difficulties (see Laird et al, 2017), they are associated with improved engagement and increased retention. Child-specific activities were more likely to occur outside of the home, either at a project's locale or as part of a social activity (Mitchell and Burgess, 2009).

When asked about the key attributes of effective outreach workers, service users referred to their being easy to talk to, straightforward, fostering positivity and hope, providing consistent support, going above and beyond, resolving problems alongside parents at their pace (particularly helping with navigating professional language, systems and processes), and operating under an ethics of care (i.e. showing respect and solidarity, while being solution focussed and asset-driven) (Gentile and Clapton, 2021).

Service users reported that having a family outreach worker who listened to them and their concerns resulted in improved communication between family members, and parents felt more confident which, in turn, helped in setting boundaries and routines for their children, while facilitating spending more time with children and promoting a parent's own active listening (Tillard, 2017; Gentile and Clapton, 2021). Parents reported that their support worker helped them support their children better and to make positive changes with regards to their substance use, improving quality of life, health and wellbeing by providing a space to help identify family strengths and rebuild relationships (McIntosh et al, 2006; Griesbach et al, 2008). Jansson et al (2003) found that intense family support saw a reduction in the amount of time children spent separated from their families and increased the likelihood of children being returned to their mothers' care. However, while there is evidence that the provision of an outreach worker is effective in leading to positive outcomes in terms of emotional wellbeing and support, caution should be taken with regards to the generalisability of the data due to its small sample size.

4.4.5 Peer support

One of the most effective elements identified in the evidence base in WFA services is peer support. This can take the form of involvement in peer discussion groups, and research suggests that they are seen as a safe space to converse with others with similar or shared experiences. Research also highlights the empowering effect of group interventions on individuals, restoring or increasing a feeling of agency as a result of engaging in mutual provision of support and advice (Covington, 2002; To et al, 2021). The literature suggests that involvement in peer discussion groups has a positive effect on substance use outcomes as well offering powerful therapeutic benefits across a range of wider wellbeing and social outcomes.^[6]

A qualitative mixed method evaluation of a WFA intervention in the North East of England found that participation in a facilitated peer support group helped mothers to move beyond the individualisation and self-blame attached to substance use (Thompson, 2022). Participants reported a range of positive effects, including that this allowed them to vocalise a shared experience, realise they were not alone, offer advice and support to others at different stages of family recovery, reconnect and feel a sense of community and belonging^[7], and feel safe and supported without the fear of being judged by either professional facilitators or other group members (Thompson, 2022).

Similarly, using qualitative findings from 13 evaluated Moving Parents and Children Together (MPACT) projects^[8], Templeton (2012) found that families benefited from meeting others having similar experiences to theirs and that this led to improved communication and a greater understanding of problem substance use and its impact on different members of the family. Furthermore, the research indicated that peer groups were very effective in giving children the space to meet other children 'just like them'. The study found improvements in familial relationships, a reduction in social service involvement in family life. Children were found to have improved educational performance, increased confidence and better management of negative emotions such as anger (Templeton, 2012).

Peer support was also found to assist the family as a whole in a number of salient ways in a pilot project in England (Murphy et al, 2010). Holding Families involved all members of a family in addressing parental substance use and its effects through a combination of facilitated family discussions, peer groups and individual direct support (e.g. activities with children). Using semi-structured interviews with children, parents and practitioners, the evaluation found that a whole family approach, and the practical sessions offered (mainly through group discussions), resulted in effective partnerships between parents and practitioners. Additionally, the peer discussions were found to be correlated with increased attempts to control or reduce substance use and to re-establish better parenting behaviours, including being more available to children by challenging parental isolation (Murphy et al, 2010).

An evaluation of family interventions in London (Velleman et al, 2005) for families affected by parental alcohol use, found that children became less anxious by the end of the project, coping responses improved, and in some cases this translated to improved school attendance, increased academic achievement and overall relationships with others. Parents, furthermore, reported becoming more aware of the impacts of their substance use and felt committed to reducing such impacts in the future (Velleman et al, 2005). This sat alongside improvements in self-esteem, family functioning with better communication, eating meals together, partaking in joint parent-child activities, with children reporting being able to regain a sense of 'childhood' (Copello et al, 2009; Murphy et al, 2010). Two-thirds of parents who engaged in two or more sessions saw an increase in sustained abstinence from substance use. As Copello et al (2009) found, there exists robust evidence that working with family members affected by substance use can trigger entry into treatment for the substance user and where this is not the case, there is an overall marked improvement in family outcomes that seek to address signs of distress and improve overall wellbeing.

4.5 Workforce implications

There is little literature with regards to the implications adopting a WFA has for substance use practitioners or the wider workforce. However, there are some references to workforce challenges that have arisen in practice. Implications for workforce are treated as an "add on" in the evidence base, which nonetheless acknowledges the importance of adopting specific workplace practices to encourage participation and, by extension, lead to the production of positive outcomes in relation to substance use.

4.5.1 Training

Evidence suggests that training staff to interact with service users in non-judgemental ways, acknowledging the importance of communication and challenging negative attitudes towards of substance use is important. Mitchell and Burgess (2009) note the importance of building upon practitioners' knowledge, awareness and understanding of the context in which substance use occurs. Specifically, they call for work to be done to develop practitioners' understanding of the factors that influences behaviour and family dynamics. Taylor et al (2008) argue that equipping practitioners with the skills to understand the psychological pressures families experience can assist practitioners in developing strategies for more effective and sustained engagements with families affected by substance use. As discussed above, developing positive relationships between service users and practitioners is important in ensuring engagement and, in turn, making progress (Thompson, 2020; McCarthy et al, 2020). Indeed, "for interventions to be deemed positive and something worthy of voluntary engagement, workers need to be caring and available, understand parents' problems and importantly acknowledge their strengths" (Thompson, 2022, p.633). Time, creativity and skill are required to overcome a service user's feelings of isolation or mistrust of professionals (Murphy et al, 2010).

Only one study focussed specifically on the workforce implications of adopting a WFA for substance use practitioners. Hampton (2012) evaluated a training program aimed at promoting a whole organisational shift towards a greater involvement of affected family members in treatment and services over a three year period. It found that staff who received immediate training reported increases in positive attitudes towards family-focussed practice and an increase in the adoption of family-focussed practices in their everyday work routines. The evaluation found that a large proportion of the barriers to family work could be alleviated through continued training and support, alongside changes to organisational structure and procedures. Hampton (2012) argues that family-focussed practices need to be developed over time and require a degree of flexibility from practitioners, with many finding family work complicated due to a series of factors beyond their control (e.g. time, funding and resources, and competing priorities of outside agencies).

Hampton's (2012) evaluation highlights some important insights for the implementation of whole family work. Specifically, there were a series of 'in house barriers' that were difficult to overcome. This included senior or staff with more experience expressing deep rooted preferences for individualistic practice, resulting in their contesting any institutional change. As such, the transition from individualistic to family-orientated work was found to present a difficult shift for staff, requiring close supervision (Hampton, 2012; see Jones and Scannell, 2002).

The research also noted the challenge of proving the "value" of whole family work, with there being notable resistance to such approaches in psychiatric services (Hampton, 2012). Staff members reported feeling that family work in substance use treatment was not recognised as valuable and therefore was underfunded by commissioners (Hampton, 2012). This sits alongside the ecological backdrop of health and social care, in which care is provided against a backdrop of diminishing resources, higher workloads and burnout (Gentile and Clapton, 2021).

This suggests that the implementation of WFAs requires an ideological and financial commitment to transformation (Forrester et al, 2013; O'Conner, 2013). This is noted by McCarthy et al (2020) who argue that there is a relatively short history of trauma-informed working in the UK, matched by a lack of routine staff training in trauma-informed practice (Tomkins and Neal, 2018). There is therefore a need for resource for recruitment and retention of staff, and for staff training and support – especially where care is provided 24 hours a day (Gentile and Clapton, 2021).

McCarthy et al (2020) note that training should be provided before starting work and include topics from attachment theory to motivational interviewing, and management training so managers can support staff creativity. They also recommend the establishment of groups to encourage staff to reflect upon their practice. They also emphasise the risk that staff exposure to traumatic material (e.g. the service user's stories or experiences) might have a negative effect on their own mental health, leading to "burn out". As such, they argue that it is important for trauma-informed staff to be trained in maintaining personal and professional boundaries in order to safeguard their own wellbeing.

4.5.2 Negative attitudes and stigma

Challenging any negative attitudes that might exist in the workforce – and that constitute a major barrier to engagement – needs to be a priority in skill and training development (Kroll and Taylor, 2009; O'Connor et al, 2014). Mason (2012) in their examination of the perspectives of parents using intensive family support services in England, documented an extensive body of work that focusses on how social workers talk to parents. Forrester et al (2008) found that power asymmetries between social workers and service users impacted negatively on outcomes.

In overcoming these differences, McCarthy et al (2020) advocate for the importance of having practitioners with lived experience, thus providing a common ground and "language" between servicer users and workers. They also advocate for service users to have some say over the identity characteristics of the teams they work with. For instance, it is acknowledged that women who use substances may also have experienced domestic abuse (Thompson, 2020). As such, working exclusively with men may discourage women from engaging with services or may act a trigger, leading to increased harms (e.g. substance use) and presenting significant challenges to trauma-informed practice (Thompson, 2020).

4.5.3 Inter-agency collaboration

Multi-agency co-operation and joined-up working were found to be key to achieving positive outcomes. There was a sense that services remain fragmented despite there being evidence that collaboration across services is crucial to the production of positive outcomes (Gentile and Clapton, 2021). For example, Kroll and Taylor's (2009) research into family interventions in the South West of England found that challenges included the existence of varying thresholds for interventions between collaborating services, confusion over confidentiality, the interpretation of protocols, and the definition of 'significant harm'.

Mitchell and Burgess (2009) call for the use of organisational structures to facilitate collaboration and co-ordination – streamlining roles and responsibilities, and implementing joined-up working practices. They argue that policies and procedures must be in place and reviewed on a regular basis to ensure they remain relevant and effective. Kroll and Taylor (2009) note the effectiveness of having a lead professional with specialist knowledge on drug misuse who is 'twin trained' and orchestrates the collaboration between different services.

Further emphasis was placed on the need for clarity around confidentiality and information sharing, especially at points of early intervention rather than only where child protection concerns are identified (McKeganey et al, 2002). A range of practices that can be used to assist cross-agency collaboration are suggested, including: designated advice-giving posts (e.g. child and family specialists located in drug and alcohol teams); crossover responsibilities for managers or managers across agencies; regular meetings to discuss emerging issues; or having joint commissioning frameworks (Kearney et al cited in Mitchell and Burgess, 2009).

4.6 Limitations with the existing evidence base

While the above literature shows clear examples of good practice in terms of working with families, there remains a number of identifiable caveats with regards to the reliability of the data.

4.6.1 Sample size

A large proportion, but not all (see Copello, Templeton and Velleman, 2006) of the findings in the studies were based on research undertaken with relatively small sample sizes and groups in general were not well represented. Despite this, it should be noted that the samples consulted in the various studies were generally representative of the demographics of users of a particular service. Nevertheless this does raise concerns with regards to the generalisability of the research findings. Small sample sizes mean that "what works" in the context of family support is based on the outcomes of few families and therefore questions remain about 1) whether "what works" could be effectively scaled up and 2) whether similar results would be yielded if such practices were adopted in services that catered for a greater number of families.

In addition, a large proportion of the evidence is quantitative. Kourgiantakis et al (2021) note that only five per cent of empirical studies consulted in their scoping review of family interventions in substance use contexts (95 articles) were qualitative. Despite this emphasis on quantitative data, researchers generally acknowledge the value of qualitative research on family interventions as it provides sociocultural perspectives on family needs for service provision, and helps to better understand the barriers and facilitating factors experienced by families (Copello et al, 2005; Neal et al, 2005; Orford, 2008; Kourgiantakis et al, 2021). Part of this focus on quantitative research is driven by the desire to measure progress against specific outcomes (Kourgiantakis et al, 2021). However this provides limited evidence to contribute towards an increased understanding of how services are experienced by service users and limits the scope for including families in service design, delivery and evaluation.

Finally, research that has used qualitative methods to understand family experiences of intervention practices often relies on questionnaires, or interviews where other service members, facilitators or family members were present (O'Connor et al, 2014). Methodologically, this requires service-users to self-report on improved outcomes and there may be a risk of bias in terms of individual desires or feeling indebted to services. Alternatively, interview participants may chose to withhold certain information as a result of who was present at the time of data collection. There is therefore some caution to be had when interpreting the research findings.

4.6.2 Family composition

There are, similarly, concerns with regards to the composition of families included in the research reviewed. As previously discussed, there have been attempts to broaden definitions and understandings of "the family", moving beyond nuclear representations to include a whole host of individuals not strictly biologically related. However, the research reviewed uses an array of strict exclusion criteria for their participants which results in a narrow selection of families as broadly defined by biological relatedness (see Burgess, 2011). Furthermore, the studies do not systematically involve families as a whole and provide little clarity as to individual family composition (Mason, 2012; Alderson et al, 2022).

All the research cited in this review was limited to children and their parent(s), with little reference to the involvement of other care givers (directly related or not). Interestingly, the narrow understanding of family composition at the level of service provision has been attributed to the location in which family interventions take place (Laid et al, 2017). As mentioned above, interactions with families largely occur within the home and Laird et al (2017) argue that as a result, practitioners are more likely to base their understanding of who is a member of the family on who lives within the home. This could contribute to services only working with "living-in" family members.

As a result, there is very limited evidence to assess "what works" for other families structures. No research was identified that purposefully engaged with or included separated parents, fathers living away from their children but continuing to co-parent, or the partners of people who use substance who do not partake in parenting activities. Instead, the existing research focusses on axial relationships, which is further exemplified by an empirical focus on families where the parents uses substances – only two papers flagged in the literature search documented the use of family inclusive practices where the person using substances was a young person (see Velleman et al, 2006; Coombes et al, 2009).

Furthermore, there is also a gender imbalance in the service users and research participants. Laird et al (2017) noted that while there is an acknowledgement of the role of kinship ties in helping families affected by substance use, there is a tendency within case files to continually redirect attention back to mother-child relationships. This is seen across the studies included in this review. The majority of participants are women, which results in limited data on fathers (Thompson, 2022; Gentile and Clapton, 2021). This gap is acknowledged across the majority of studies reviewed (Pander-Brick et al, 2014; Velleman et al, 2005; Mitchell and Burgess, 2009; Burgess, 2011; Leonard et al, 2018; Gentile and Clapton, 2021) and feeds into the gendered stigma attached to substance use and perceptions around parenting skills and proposed 'risk' (Thompson, 2022).

Finally, there is a lack of representation of families with neurodiverse members or families with different cultural and ethnic backgrounds, which in turn reshapes how the family unit is understood and the culturally-specific roles and responsibilities assigned to family members (Kourgiantakis et al, 2021).

There is a need for research to assess whether WFAs produce positive outcomes for families that are culturally diverse, composed of multiple family members with differing identities (including LGBTQ+ family relationships) and families with disabilities. All the papers included in this review explore projects undertaken within urban areas – research similarly needs to be undertaken that explores appropriate practices used to engaged isolated communities and challenges the increased visibility of families accessing services (Kroll and Taylor, 2009). Without understanding what works for different types of families, evidence on the effectiveness of WFAs should be considered with caution.