The misuse of alcohol is of major national and international public health concern and can bring significant harm and burden to family members, resulting in increased use of health and social care services and a drain on their resources. There is increasing evidence of a variety of ways of working with families affected by alcohol misuse to achieve positive outcomes. Whilst several systematic reviews have focused on interventions for people with alcohol problems, few reviews have attempted to synthesise findings from studies of family focused interventions; those which have been done are quite old, and have tended to focus on drug problems (or drug and alcohol problems), and towards outcomes related to treatment engagement and retention and the alcohol or drug consumption of the misuser. The evidence in this area is growing but is disparate and needs reviewing and synthesising in a systematic way so that informed decisions can be made about developing policy, practice and research in this area. An up-to-date review of interventions for family members would help guide the current significant and overdue shift towards more holistic treatment.
Therefore, the aims of this review were:
- To undertake a comprehensive and systematic review of the literature on psychological interventions for families affected by alcohol misuse.
- To assess the available evidence on the effectiveness of psychological interventions in improving life for family members of alcohol misusers.
The topic was both too diverse and un-developed for a very specific review to be undertaken, so a broad review methodology which followed the core steps and principles of a systematic review was adopted. Hence, broad definitions for population, type of study, intervention, outcomes and alcohol misuse were agreed to allow for as full a review as possible to be completed. There were no date, language or time restrictions. Several electronic and grey literature resources were searched. In addition, Addiction Abstracts from 1990 was hand-searched, reference lists and websites were scanned and other publications and resources known to the project team were searched. Two reviewers screened all titles and abstracts to come from the searching process; studies were combined where more than one publication related to the same study or the same sample. Two reviewers then independently assessed the eligibility and quality of all potential studies. Given the heterogeneity across the included studies, and that this was the first review in this specific area, quantitative analysis or meta-analysis was not possible; hence, a thematic and narrative approach to analysis was employed. The included studies were grouped and reviewed according to whether or not the family member received an intervention with or without the alcohol misuse.
Following screening, 88 potential publications from 79 studies were identified. Following eligibility and quality assessment, 43 publications from 34 studies were included in the review. There were 45 excluded publications (45 studies). The main reasons for exclusion were that the study was descriptive only, did not consider or include enough focus on alcohol, lacked focus on family members or was incomplete/ongoing and hence data were not available.
All included studies were in English, covered the period 1979-2009 and were mostly published articles in academic journals. Nearly half of the studies were from the USA, with a further 10 conducted in England, 5 in Australia, 3 in Sweden, 1 in Canada and 1 in Italy. Female (generally spouses/partners) and ‘white’ family members dominated. Whilst exact information was lacking in some studies, the first group of studies (where the user was not involved) included approximately 1,500 family members, and the second group of studies (where the user was involved) included around 1,200 family members.
21 studies of interventions which involve family members without the alcohol misuser
Over half of this group of studies came from the USA or the UK. Of the 15 studies which involved partners or spouses (usually female), nine were trials, involving randomisation to usually two or three groups. A range of individually oriented or group interventions were considered. Two intervention models dominated in this group, the Australian ‘Pressures to Change’ approach and the UK ‘5-step intervention’. Follow-up data were usually collected immediately post-intervention or at 12 months. Some studies were feasibility studies of a new intervention where, the addition of qualitative data made a valuable contribution.
The premise behind this group of studies is that they are oriented towards assisting family members in their own right. However, the extent to which the studies achieved this varied. Coupled with methodological limitations in many of the studies, a summary of quantitative findings is hard. Nonetheless, collectively, the studies indicated that interventions which target family members’ own needs can result in positive change across a range of domains including, for example, health, coping, stress/distress, hardship and relationship satisfaction. Where qualitative evidence was also available there was greater understanding of how the intervention can help family members as well as evidence that intervening in this way can bring benefits for others in the family group, for example children or the individual with the alcohol problem. Treatment groups generally fared better than control groups but, where interventions were compared with each other, generally no one intervention stood out as being better than another. Working with family members of alcohol misusers in these ways is in its infancy but, nonetheless, the results in this group of studies demonstrate potential.
The other six studies in this group focused on children, usually ‘adult children of alcoholics’. This group of studies was limited by their recruitment of groups of children, not all of whom were living with parental alcoholism, and by not fully reporting on sub-analyses of the group of children who did self-identify as living with parental alcoholism. In general, however, these studies report the benefits of their interventions, and offer additional ideas about potential mediating factors which, if altered, can facilitate more positive outcomes in a range of areas.
13 studies of interventions which involved family members with the alcohol misuser
The majority of the studies in this group (nine studies) involved couples therapy, primarily Behavioural Couples Therapy (BCT) or forms thereof, and were from the USA (with one Canadian study). The majority of these studies were generally well conducted, pragmatic and quantitative trials. Summative conclusions from these studies strongly indicated that couples therapy results in positive outcomes, particularly in drinking behaviour and marital adjustment, which can be maintained in varying degrees through follow-up.
The other four studies were small scale evaluations of holistic family interventions, all from the UK and including considerable qualitative data. All studies included children and young people in the study, including direct data collection from them. The studies reported changes for families in the areas of marital and family satisfaction, alcohol dependence, family dynamics, communication, family relationships, health and coping. Further, all studies presented useful evidence that adopting a therapeutic approach which closely considers family strengths and values, and how individual and familial resilience can be improved, can make a major contribution to the success of the intervention.
This review was undertaken to provide a timely and broad overview for a field in the midst of significant evolution. The review could have simply focused on a tiny number of double-blind randomised controlled trials. However, a broader approach has brought a fuller understanding of a field where few reviews have been conducted. Moreover, whilst a more limited or focused review might have increased study homogeneity, such a review would not have been able to offer such a broad and useful overview of the field, how it is developing or its strengths and limitations. A diverse range of studies and interventions was therefore identified which highlights the chronology of how this field has moved from primarily focusing on how family members can engage and support the user through treatment to adopting a wider holistic focus which considers the needs of family members in their own right. Unilateral interventions for family members and group interventions for families have become more popular. However, despite this shift many studies retained a greater focus on the person with the alcohol problem.
Population: The majority of the included studies involved adult female family members, usually spouses or partners. However, more recent work, particularly in the UK, Australia and Sweden has successfully engaged a wider range of family members albeit in quite small numbers.
Study design: Quantitative and controlled trial designs, usually conducted in the USA and involving couples therapies, dominated. Pilot and feasibility studies, which often included qualitative data, and were more likely to be unpublished, had a greater focus towards newer areas of intervention, such as unilateral and family group interventions.
Intervention: Several interventions dominated in the review, including Behavioural Couples Therapy, Pressures to Change and the 5-step Method. Where several studies considered the same intervention, a broader exploration of the associated strengths and limitations is possible.
Outcomes: Whilst a range of outcome measures was used in the included studies, there continued to be a narrow range of outcomes considered, usually related to the alcohol consumption of the misuser or marital functioning. Assessing physical and psychological well-being and coping emerged across the studies as important domains to study in terms of exploring positive change for family members in their own right.
Whilst a detailed analysis of study outcomes in each group was not undertaken, the fact that positive change was seen across the included studies (either through statistically significant analyses or thematic qualitative findings) suggests that there may not be a ‘best’ intervention for family members living with alcohol misuse. Rather, it may be that elements of support non-specific to the intervention approach itself are equally important in contributing to positive outcomes. More research is needed to better understand the active ingredients of interventions, the processes and mechanisms by which they effect change and whether other factors contribute to positive change.
Identification of studies: The deliberately broad definitions used for this review, and the use of multiple searching resources with different methods for searching, made the identification and selection of studies difficult but comprehensive. However, some studies we would have expected to identify were not identified.
Review Tools: The broad nature of the review meant that suitable tools for screening articles, assessing their quality and extracting data were not available. The authors therefore reviewed existing tools which are available for full systematic reviews and then used elements from these to develop their own tools for this review, tools which could be usefully developed and made available for other such reviews.
Quality assessment: Assessing the quality of the included studies identified variation across different aspects of the studies. Descriptions of background, design and interventions were usually adequate and the discussions were usually relevant. Studies tended to lack detail (where appropriate) in many of the same areas. However, in many cases the work being done was innovative, and the studies were pilot/feasibility studies, with little previous work on which to base decisions on things like study design. As such it is less helpful to report on those things which are usually the mainstay of randomised controlled trials and reviews published in the Cochrane Library and it was therefore necessary to broaden the assessment of quality to ensure that it was applicable to the range of studies included.
Follow-up: Another limitation of many studies related to the length of follow-up, up to a maximum of two to three years for a handful of studies. It is important that future research is able to include longer follow-ups to enhance understanding of whether change associated with family focused interventions is sustained, or results in subsequent and delayed positive change. ‘Recovery’ for a family can continue for years after the cessation of consumption, with change in different domains occurring at different time periods, so work which considers a range of outcomes over longer time periods would also be beneficial.
Implications for Research and Practice
There are several implications which inform where future practice and research in this area should be directed, and where more focused reviews could be undertaken. Firstly, there is a need to consider the involvement of a wider range of family members, and of how family members can directly or indirectly benefit from a family member receiving help. Next, given that developments in the field are still in their infancy, flexibility and innovation in service and intervention design should be encouraged. The research and development agenda, and reviews thereof, must be wider-ranging, including pilot and feasibility studies; adequately powered and conducted randomised controlled trials with sufficient follow-up; evaluations and effectiveness trials of the implementation of interventions into routine practice; and consideration of the economic benefits of family interventions. Qualitative methods are a necessary part of all areas of this agenda. Finally, whilst many of the studies reviewed here (perhaps because of the dominance of work from the USA) are based on a diagnostic definition of alcoholism, and hence on a treatment goal of abstinence for the drinker; future work could usefully extend this to consider how interventions might also operate when broader definitions and goals, such as ‘problem drinking’ and moderate consumption, are considered. Further work would also be helpful to consider the impact of the severity of the alcohol problem on the outcome of the intervention.
The way in which the aforementioned recommendations need to be applied will differ according to the intervention under consideration. For many interventions there are indications of promise suggesting that an applied programme of research is needed. Others already have stronger evidence in their favour, including, for example, Behavioural Couples Therapy, the 5-step approach and the Pressures to Change model, and the focus here needs to be on considering how best to roll-out the intervention into routine practice or in how to adapt an intervention to reach different sub-populations of family members or to be delivered in different settings. Despite major developments at a strategic level in some countries, and the fact that many of the included studies in this review alluded in their discussion to a need to translate the interventions studied into routine practice, there was often little discussion on how to do this, although some of the challenges in doing this have been highlighted. Some work offers ideas and examples on how to successfully bridge this gulf between research and practice, the opportunities which arise and the barriers which need to be overcome.
In conclusion, the evidence is mounting that involving the family in treatment, including responding to their own needs directly, and largely regardless of the nature of the intervention, can bring huge benefits to family members, including children and the alcohol misusing relatives, as well as bringing potential cost-savings to services.
The authors would like to thank the AERC for funding the review. Through their membership of the project steering group or specific support, thanks are also extended to Ray Hodgson, David Foxcroft, Gordon Taylor, Isobel Stark and Justin Hodds
Templeton L, Velleman R & Russell C (2010). Psychological Interventions with Families of Alcohol Misusers: A Systematic Review. Addiction Research & Theory 18(6): 616-648.“
Lorna Templeton, Richard Velleman & Claire Russell
Mental Health Research & Development Unit (Avon & Wiltshire Mental Health Partnership NHS Trust and the University of Bath), 22-23 Eastwood, University of Bath, BA2 7AY