6.1.1. Current practice

Services for people who are alcohol dependent and harmful drinkers are commonly delivered by statutory and non-statutory providers. The field is undergoing rapid change across different areas of the country due to the impact of the commissioning process. Traditionally, services have been provided by teams where the detoxification and counselling aspects of treatment have been fairly clearly separated. Within the NHS, teams tend to consist of different disciplines including nurses, counsellors, medical practitioners and, less often, other professions such as psychologists and occupational therapists. Teams are commonly under-resourced with practitioners having high caseloads and limited access to supervision. Most practice involves an eclectic approach that combines strategies from various psychological approaches. A more recent development involves contracts between commissioners and providers that may determine, for example the number of sessions to be delivered, yet this is rarely informed by the evidence and tends to be driven by pragmatic or resource issues (Drummond et al., 2005).

Whilst the research literature to date has concentrated mostly on the comparison of well-defined treatment interventions commonly incorporated into treatment manuals, this stands in contrast to what is normally delivered in routine practice. Despite the research on psychological treatments, current UK practice is not underpinned by a strong evidence base and there is wide variation in the uptake and implementation of psychological approaches to treatment across services (Drummond et al., 2005).

A number of factors may contribute to the low implementation of evidence-based psychological interventions. First, there is a lack of availability of reviews of the current evidence in a clear and practical format that can be accessible to practitioners, managers and commissioners. This has led to a weak dissemination of the evidence base concerning psychological interventions for alcohol misuse within routine service provision. Second, there is the varied composition of the workforce with a range of training experiences, not all of which include training in the delivery of psychological interventions. Furthermore, as noted by Tober and colleagues (2005), training programmes for the management of substance misuse vary widely in content with no consensus on methods to provide and evaluate such training or to maintain its effects. Supervision of psychological interventions is equally varied and not always available. Finally, there is a tendency in the field to eclecticism fuelled by the perception that all approaches are either equally valid or equally ineffective.

6.2.1. The therapeutic alliance

There are various definitions of the therapeutic alliance, but in general terms it is viewed as a constructive relationship between therapist and client, characterised by a positive and mutually respectful stance in which both parties work on the joint enterprise of change. Bordin (1979) conceptualised the alliance as having three elements: agreement on the relevance of the tasks (or techniques) employed in therapy; agreement about the goals or outcomes the therapy aims to achieve; and the quality of the bond between therapist and patient.

There has been considerable debate about the importance of the alliance as a factor in promoting change, with some commentators arguing that technique is inappropriately privileged over the alliance, a position reflected in many humanistic models where the therapeutic relationship itself is seen as integral to the change process, with technique relegated to a secondary role (for example, Rogers, 1951). The failure of some comparative trials to demonstrate differences in outcome between active psychological therapies (for example, Elkin, 1994; Miller & Wilbourne, 2002) is often cited in support of this argument and is usually referred to as ‘the dodo-bird hypothesis’ (Luborsky et al., 1975). However, apart from the fact that dodo-bird findings may not be as ubiquitous as is sometimes claimed, this does not logically imply that therapy technique is irrelevant to outcome. Identifying and interpreting equivalence of benefit across therapies remains a live debate (for example, Ahn & Wampold, 2001; Stiles et al., 2006) but should also include a consideration of cost effectiveness as well as clinical efficacy (NICE, 2008a).

Meta-analytic reviews report consistent evidence of a positive association of the alliance with better outcomes with a correlation of around 0.25 (for example, Horvath & Symonds, 1991; Martin et al., 2000), a finding that applies across a heterogeneous group of trials (in terms of variables such as type of therapy, nature of the disorder, client presentation, type of measures applied and the stage of therapy at which measures are applied). However, it is the consistency rather than the size of this correlation that is most striking because a correlation of 0.25 would suggest it could account for only 6% of the variance in the outcome. Specific studies of the role of the alliance in drug and alcohol treatment programmes have been conducted. Luborsky and colleagues (1985), Connors and colleagues (1997) and Ilgen and colleagues, (2006) reported a relationship between treatment outcomes, but others (for example, Ojehagen et al., 1997) have not. Ojehagen and colleagues (1997) suggest that this discrepancy between the various studies may have arisen from methodological differences between the studies; in contrast to Luborsky and colleagues (1985), Connors and colleagues (1997) and Ilgen and colleagues (2006), in Ojehagen and colleagues' (1997) study ratings of the alliance were made by an independent rater from video tapes as opposed to ratings made by the therapist early in treatment. This is consistent with other studies; for example, Feeley and colleagues (1999) reported that alliance quality was related to early symptom change. Therefore, it seems reasonable to debate the extent to which a good alliance is necessary for a positive outcome of an intervention, but it is unlikely to be sufficient to account for the majority of the variance in outcome.

6.2.2. Therapist competence

Studies of the relationship between therapist competence and outcome suggest that all therapists have variable outcomes, although some therapists produce consistently better outcomes (for example, Okiishi et al., 2003). There is evidence that more competent therapists produce better outcomes (Barber et al., 1996 and 2006; Kuyken & Tsivrikos, 2009). This is also the case for psychological interventions in the alcohol field; the Project MATCH Research Group (1998) reported therapist differences that impact on outcome. A number of studies have also sought to examine more precisely therapist competence and its relation to outcomes; that is, what is it that therapists do to achieve good outcomes? A number of studies are briefly reviewed here.

This section draws on a more extensive review of the area by Roth and Pilling (2011), which focused on CBT because this area had the most extensive research. In an early study, Shaw and colleagues (1999) examined competence in the treatment of 36 patients treated by eight therapists offering CBT as part of the National Institute of Mental Health trial of depression (Elkin et al., 1989). Ratings of competence were made on the Cognitive Therapy Scale (CTS). Although the simple correlation of the CTS with outcome suggested that it contributed little to outcome variance, regression analyses indicated a more specific set of associations; specifically, when controlling for pre-therapy depression scores, adherence and the alliance, the overall CTS score accounted for 15% of the variance in outcome. However, a subset of items on the CTS accounted for most of this association.

Some understanding of what may account for this association emerges from three studies by DeRubeis's research group (Feeley et al., 1999; Brotman et al., 2009). All of the studies made use of the Collaborative Study Psychotherapy Rating Scale (CSPRS: Hollon et al., 1988), subscales of which contained items specific to CBT. On the basis of factor analysis, the CBT items were separated into two subscales labelled ‘cognitive therapy – concrete’ and ‘cognitive therapy – abstract’. Concrete techniques can be thought of as pragmatic aspects of therapy (such as establishing the session agenda, setting homework tasks or helping clients identify and modify negative automatic thoughts). Both DeRubeis and Feeley (1990) and Feeley and colleagues (1999) found some evidence for a significant association between the use of ‘concrete’ CBT techniques and better outcomes. The benefits of high levels of competence over and above levels required for basic practice has been studied in most detail in the literature on CBT for depression. In general, high severity and comorbidity, especially with Axis II pathology, have been associated with poorer outcomes in therapies, but the detrimental impact of these factors is lessened for highly competent therapists. DeRubeis and colleagues (2005) found that the most competent therapists had good outcomes even for patients with the most severe levels of depression. Kuyken and Tsivrikos (2009) found that therapists who are more competent have better patient outcomes regardless of the degree of patient comorbidity. In patients with neurotic disorders (Kingdon et al., 1996) and personality disorders (Davidson et al., 2004), higher levels of competence were associated with greater improvements in depressive symptoms. Although competence in psychological therapies is hard to measure in routine practice, degrees of formal training (Brosan et al., 2007) and experience in that modality (James et al., 2001) are associated with competence and are independently associated with better outcomes (Burns & Nolen-Hoeksema, 1992). All therapists should have levels of training and experience that are adequate to ensure a basic level of competence in the therapy they are practicing, and the highest possible levels of training and experience are desirable for those therapists treating patients with severe, enduring or complex presentations. In routine practice in services providing psychological therapies for depression, therapists should receive regular supervision and monitoring of outcomes. Roth and colleagues (2010) reviewed the training programmes associated with clinical trials as part of a programme exploring therapist competence (Roth & Pilling, 2008). They showed that clinical trials are associated with high levels of training, supervision and monitoring–factors that are not always found in routine practice. This is partly due to the inadequate description of training programmes in the trial reports. However, there is an increasing emphasis on describing the process of training in clinical trials, the report by Tober and colleagues (2005) being a notable recent publication describing the training programme for UKATT.

Trepka and colleagues (2004) examined the impact of competence by analysing outcomes in Cahill and colleagues' (2003) study. Six clinical psychologists (with between 1 and 6 years' post-qualification experience) treated 30 clients with depression using CBT, with ratings of competence made on the CTS. In a completer sample (N = 21) better outcomes were associated with overall competence on the CTS (r = 0.47); in the full sample this association was only found with the ‘specific CBT skills’ subscale of the CTS. Using a stringent measure of recovery (a Beck Depression Inventory score no more than one SD from the non-distressed mean), nine of the ten completer patients treated by the more competent therapists recovered, compared with four of the 11 clients treated by the less competent therapists. These results remained even when analysis controlled for levels of the therapeutic alliance.

Miller and colleagues (1993) looked at therapist behaviours in a brief (two-session) ‘motivational check-up’; they identified one therapist behaviour (a confrontational approach) that was associated with increased alcohol intake. Agreeing and monitoring homework is one of the set of ‘concrete’ CBT skills identified above. All forms of CBT place an emphasis on the role of homework because it provides a powerful opportunity for clients to test their expectations. A small number of studies have explored whether compliance with homework is related to better outcomes, although rather fewer have examined the therapist behaviours associated with better client ‘compliance’ with homework itself. Kazantzis and colleagues (2000) report a meta-analysis of 27 trials of cognitive and/or behavioural interventions that contained data relevant to the link between homework assignment, compliance and outcome. In 19 trials, clients were being treated for depression or anxiety; the remainder were seen for a range of other problems. Of these, 11 reported on the effects of assigning homework in therapy and 16 on the impact of compliance. The type of homework varied, as did the way in which compliance was monitored, although this was usually by therapist report. Overall there was a significant, although modest, association between outcome and assigning homework tasks (r = 0.36), and between outcome and homework compliance (r = 0.22). While Kazantzis and colleagues (2000) indicate that homework has greater impact for clients with depression than anxiety disorders, the number of trials on which this comparison is made is small and any conclusions must therefore be tentative.

Bryant and colleagues (1999) examined factors leading to homework compliance in 26 clients with depression receiving CBT from four therapists. As in other studies, greater compliance with homework was associated with better outcome. In terms of therapist behaviours, it was not so much therapists' CBT-specific skills (such as skilfully assigning homework or providing a rationale for homework) that were associated with compliance, but ratings of their general therapeutic skills and particularly whether they explicitly reviewed the homework assigned in the previous session. There was also some evidence that compliance was increased if therapists checked how the client felt about the task being set and identified potential difficulties in carrying it out.

6.5.1. Aim of review

This section aims to review the evidence for psychological interventions without pharmacological interventions for the treatment of alcohol dependence and harmful alcohol use. The literature reviewed in this section is focused on a reduction or cessation of drinking and hence assesses any outcomes pertaining to this. Most of the literature in the field is focused on adults over the age of 18 years. However, for young people under the age of 18 years old, literature assessing the clinical efficacy of psychological therapies for alcohol misuse alone (without comorbid drug misuse) is limited. The psychological evidence below is for an adult population only and a review of the evidence for the treatment of young people is described in Section 6.22.

Psychological interventions were considered for inclusion in the review if they were:

• Planned treatment
• For treatment-seeking participants only (of particular importance for the brief interventions because the scope did not cover opportunistic brief interventions – see scope, Appendix 1)
• Manual-based or, in the absence of a formal manual, the intervention should be well-defined and structured
• Ethical and safe

The following psychological therapies used in the treatment of alcohol misuse were considered for inclusion in this guideline:

• Brief interventions (planned only)

  –

  for example, psychoeducational and motivational techniques

• Self-help based treatments

  –

  brief self-help interventions (including guided self-help/bibliotherapy)

• TSF
• Cognitive behavioural-based therapies

  –

  standard cognitive behavioural therapy

  –

  coping skills

  –

  social skills training

  –

  relapse prevention

• Behavioural therapies

  –

  cue exposure

  –

  BSCT

  –

  contingency management

  –

  aversion therapy

• MET
• Social network and environment-based therapies

  –

  social behaviour and network therapy (SBNT)

  –

  the community reinforcement approach

• Counselling

  –

  couples therapy (including behavioural couples therapy and other variants of couples therapy)

• Family-based interventions

  –

  functional family therapy

  –

  brief strategic family therapy

  –

  multisystematic therapy

  –

  five-step family interventions

  –

  multidimensional family therapy

  –

  community reinforcement and family training

• Psychodynamic therapy

  –

  short-term psychodynamic intervention

  –

  supportive expressive psychotherapy.

In addition, physical therapies such as meditation and acupuncture are also covered in this review.

Good quality RCT evidence for the clinical efficacy of some of the psychological therapies listed was not always available. Therefore, the evidence summaries in this chapter describe the psychological therapies for which evidence of sufficient quality (see Chapter 3 for methodological criteria) was available. There are a number of useful studies that add value to the RCT data presented and they are included in this review. For the purpose of this guideline, and to obtain an overview of the available literature, studies that have met other methodological criteria are described in the evidence summaries of the individual therapies.

Full characteristics of included studies, forest plots and GRADE profiles can be found in Appendix 16d, 17c and 18c, respectively, because they were too extensive to place within this chapter.

6.5.2. Review questions

Primary review questions addressed in this chapter:

1. For people with alcohol dependence or who are harmful drinkers, is psychological treatment x, when compared with y, more clinically and cost effective and does this depend on:
   • presence of comorbidities
   • subtypes (matching effects)
   • therapist-related factors (quality, therapeutic alliance, competence, training and so on).

6.7.1. Definition

Motivational enhancement therapy (MET) is the most structured and intensive motivational-based intervention. It is based on the methods and principles of motivational interviewing (Miller et al., 1992). It is patient centred and aims to result in rapid internally-motivated changes by exploring and resolving ambivalence towards behaviour. The treatment strategy of motivational interviewing is not to guide the client through recovery step-by-step, but to use motivational methods and strategies to utilise the patient's resources. A more specific manualised and structured form of motivational interviewing based on the work of Project MATCH is usually utilised (Project MATCH Research Group, 1993).

Brief motivational interventions include the computerised Drinker's Check-Up (DCU), which assesses symptoms of dependence, alcohol-related problems and motivation for change, and ‘feedback, responsibility, advice, menu, empathy, self-efficacy’ (FRAMES; Bien et al., 1993³²).

6.7.2. Clinical review protocol (motivational techniques)

Information about the databases searched and the inclusion/exclusion criteria used for this section of the guideline can be found in Chapter 3 (further information about the search for health economic evidence can be found in Section 6.21 of this chapter). See Table 32 below for a summary of the clinical review protocol for the review of motivational techniques.

Table 32

Clinical review protocol for the review of motivational techniques.

6.7.3. Studies considered for review³³

The review team conducted a systematic review of RCTs that assessed the beneficial or detrimental effects of motivational techniques in the treatment of alcohol dependence or harmful alcohol use. See Table 33 for a summary of the study characteristics. It should be noted that some trials included in analyses were three- or four-arm trials. To avoid double counting, the number of participants in treatment conditions used in more than one comparison was divided (by half in a three-arm trial and by three in a four-arm trial).

Table 33

Summary of study characteristics for motivational techniques.

Eight trials relating to clinical evidence met the eligibility criteria set by the GDG, providing data on 4,209 participants. All eight studies were published in peer-reviewed journals between 1997 and 2007. A number of studies identified in the search were initially excluded because they were not relevant to this guideline. Studies were excluded because they did not meet methodological criteria (see Chapter 3). When studies did meet basic methodological inclusion criteria, the main reason for exclusion was not meeting the drinking quantity/diagnostic criteria; that is, participants were not drinking enough to be categorised as harmful or dependent drinkers, or less than 80% of the sample met criteria for alcohol dependence or harmful alcohol use. Other reasons were that treatment was opportunistic as opposed to planned, the study was not directly relevant to the review questions, or no relevant alcohol-focused outcomes were available. A list of excluded studies can be found in Appendix 16d.

6.7.4. Evidence summary³⁴

The GRADE profiles and associated forest plots for the comparisons can be found in Appendix 18c and Appendix 17c, respectively.

Motivational techniques versus minimal intervention control

One computerised session of MET (drinker's check up) was significantly better than control in reducing average drinks per day at 1-month follow-up (moderate effect size). However, this finding was based on the results of a single study. Furthermore, no significant difference in average drinks per day and DDD was observed between the drinker's check up and control at 2- and 12-month follow-up.

MET (with relapse prevention) (ROSENBLUM2005b) was significantly more effective than control at reducing heavy alcohol use when assessed at 5-month follow-up (moderate effect size). This was further supported by the SELLMAN2001 study, which favoured MET over control in the number of people who drank excessively and frequently (ten or more drinks, six or more times) at 6-month follow-up (large effect size). However, this effect was not observed at long follow-up assessment (5 years). Although no significant difference was observed between groups in reducing the days on which any alcohol was consumed, the analyses showed a trend favouring MET with relapse prevention over control (p = 0.07). No significant difference in attrition rates were observed between MET and control groups across studies.

The quality of this evidence is moderate and further research is likely to have an important impact on confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 34.

Table 34

Motivational techniques versus control evidence summary.

Motivational techniques versus other active intervention

The clinical evidence showed that no significant difference could be found between motivational techniques and other active interventions in maintaining abstinence at up to 15-month follow-up. Furthermore, no difference between groups was observed in reducing the number of participants who had lapsed or reducing heavy drinking at all follow-up points.

Other therapies (namely CBT and TSF) were more effective than motivational techniques in reducing the quantity of alcohol consumed when assessed post-treatment. However, the effect size was small (0.1) and was no longer seen at longer follow-up points of 3 to 15 months.

No significant difference was observed between groups in attrition rates post-treatment or at 3-month follow-up. However, other therapies were more effective at retaining participants at 6-month follow-up (low effect size). Follow-up periods longer than 6 months did not indicate any significant difference between groups.

The quality of this evidence is moderate, therefore further research is likely to have an important impact on confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 35.

Table 35

Motivational techniques versus other intervention evidence summary.

6.8.1. Definition

TSF is based on the 12-step or AA concept that alcohol misuse is a spiritual and medical disease (see Section 6.4 for a discussion of AA). As well as a goal of abstinence, this intervention aims to actively encourage commitment to and participation in AA meeting. Participants are asked to keep a journal of AA attendance and participation, and are given AA literature relevant to the ‘step’ of the programme that they have reached. TSF is highly structured and manualised (Nowinski et al., 1992) and involves a weekly session in which the patient is asked about their drinking, AA attendance and participation, given an explanation of the themes of the current sessions, and goals for AA attendance are set.

6.8.2. Clinical review protocol (12-step facilitation)

Information about the databases searched and the inclusion/exclusion criteria used for this section of the guideline can be found in Chapter 3 (further information about the search for health economic evidence can be found in Section 6.21). See Table 36 below for a summary of the clinical review protocol for the review of TSF.

Table 36

Clinical review protocol for the review of 12-step facilitation.

6.8.3. Studies considered for review

The review team conducted a systematic review of RCTs that assessed the beneficial or detrimental effects of TSF in the treatment of alcohol dependence or harmful alcohol use. See Table 37 for a summary of the study characteristics. It should be noted that some trials included in analyses were three- or four-arm trials. To avoid double counting, the number of participants in treatment conditions used in more than one comparison was divided (by half in a three-arm trial, and by three in a four-arm trial).

Table 37

Summary of study characteristics for 12-step facilitation.

Six trials relating to clinical evidence met the eligibility criteria set by the GDG, providing data on 2,556 participants. All six studies were published in peer-reviewed journals between 1997 and 2009. A number of studies identified in the search were initially excluded because they were not relevant to this guideline. Studies were excluded because they did not meet methodological criteria (see Chapter 3). When studies did meet basic methodological inclusion criteria, the main reason for exclusion was that the studies were assessing the efficacy of 12-step groups (that is, AA) directly (not TSF) and hence were also naturalistic studies. Other reasons included a drug and not alcohol focus, secondary analysis and not being directly relevant to the current guideline. A list of excluded studies can be found in Appendix 16d.

6.8.4. Evidence summary

The GRADE profiles and associated forest plots for the comparisons can be found in Appendix 18c and Appendix 17c, respectively.

12-step facilitation versus other active interventions

The clinical evidence revealed no significant difference between TSF and other active interventions in maintaining abstinence, reducing heavy drinking episodes when assessed post-treatment and at various follow-up points up to 12 months. TSF was significantly better than other active interventions in reducing the amount of alcohol consumed when assessed at 6-month follow-up. However, the effect size was small (SMD = −0.09) and no significant difference between groups was observed for any other follow-up points.

No significant difference in attrition rates was observed between TSF and other active interventions in attrition post-treatment and up to 6-month follow-up. However, those receiving TSF were more likely to be retained at 9-month follow-up, although his difference was not observed at 12- and 15-month follow-up.

The quality of this evidence is high, therefore further research is unlikely to change confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 38.

Table 38

12-Step Facilitation versus other intervention evidence summary.

Comparing different formats of 12-step facilitation

Directive TSF was more effective at maintaining abstinence than motivational TSF up to 12-month follow-up (RR = −0.41 to −0.81 across follow-up points). However, no difference between groups was observed in reducing heavy drinking episodes.

In addition, intensive TSF was significantly more effective than standard TSF in maintaining abstinence at 12-month follow-up (RR = 0.81).

No significant difference between TSF methods was observed in attrition post-treatment or at various follow-up points up to 12 months.

Additionally, KAHLER2004 was identified as assessing brief advice to facilitate AA involvement versus a motivational enhancement approach to facilitate AA involvement. This study could not be included in analyses because data could not be extracted. However, the study reported that although AA attendance was associated with better drinking outcomes, the more intensive motivational enhancement format of facilitating involvement did not improve involvement in AA and hence did not result in better alcohol outcomes.

The quality of this evidence is moderate, therefore further research is likely to have an important impact on confidence in the estimate of the effect and may change the estimate (see Appendix 18c). An evidence summary of the results of the meta-analyses can be seen in Table 39.

Table 39

Comparing different formats of 12-step facilitation evidence summary.

6.9.1. Definition

CBT encompasses a range of therapies, in part derived from the cognitive behavioural model of affective disorders in which the patient works collaboratively with a therapist using a shared formulation to achieve specific treatment goals. Such goals may include recognising the impact of behavioural and/or thinking patterns on feeling states and encouraging alternative cognitive and/or behavioural coping skills to reduce the severity of target symptoms and problems. Cognitive behavioural therapies include standard CBT, relapse prevention, coping skills and social skills training.

6.9.2. Clinical review protocol (cognitive behavioural therapies)

Information about the databases searched and the inclusion/exclusion criteria used for this Section of the guideline can be found in Chapter 3 (further information about the search for health economic evidence can be found in Section 6.21). See Table 40 below for a summary of the clinical review protocol for the review of cognitive behavioural therapies.

Table 40

Clinical review protocol for the review of cognitive behavioural therapies.

6.9.3. Studies considered for review

The review team conducted a systematic review of RCTs that assessed the beneficial or detrimental effects of cognitive behavioural therapies in the treatment of alcohol dependence or harmful alcohol use. See Table 41 for a summary of the study characteristics. It should be noted that some trials included in analyses were three- or four-arm trials. To avoid double counting, the number of participants in treatment conditions used in more than one comparison was divided (by half in a three-arm trial, and by three in a four-arm trial).

Table 41

Summary of study characteristics for cognitive behavioural therapies.

Twenty RCT trials relating to clinical evidence met the eligibility criteria set by the GDG, providing data on 3,970 participants. All 20 studies were published in peer-reviewed journals between 1986 and 2009. A number of studies identified in the search were initially excluded because they were not relevant to this guideline. Studies were excluded because they did not meet methodological criteria (see Chapter 3). When studies did meet basic methodological inclusion criteria, the main reasons for exclusion were not having alcohol-focused outcomes that could be used for analysis and not meeting drinking quantity/diagnosis criteria (that is, participants were not drinking enough to be categorised as harmful or dependent drinkers, or less than 80% of the sample met criteria for alcohol dependence or harmful alcohol use). Other reasons were that the study was outside the scope of this guideline, presented secondary analyses, and was drug-focused or did not differentiate between drugs and alcohol, and was focused on aftercare. A list of excluded studies can be found in Appendix 16d.

6.9.4. Evidence summary

The GRADE profiles and associated forest plots for the comparisons can be found in Appendix 18c and Appendix 17c, respectively.

Cognitive behavioural therapies versus treatment as usual or control

Cognitive behavioural therapies were significantly better than control at reducing heavy drinking episodes but no significant difference between groups was observed for a reduction in days any alcohol is used (assessed post-treatment) or the number of participants who have lapsed and relapsed (assessed at 3-month follow-up) when compared with treatment as usual. However, resulting in a moderate effect size, cognitive behavioural therapies were significantly better than treatment as usual in reducing the number of participants who lapsed and relapsed when assessed at 6-month follow-up. No difference between groups was observed in attrition rates post-treatment or at 6-month follow-up.

The quality of this evidence is moderate, therefore further research is likely to have an important impact on confidence in the estimate of the effect and may change the estimate (see Appendix 18c for full GRADE profile).

One study assessing cognitive behavioural therapies versus control could not be added to the meta-analyses. Källmén and colleagues 2003 could not be included because the data was presented in an unusable format. The study reported that the control group (unstructured discussion) drank significantly less alcohol at 18-month follow-up than the group receiving coping skills. An evidence summary of the results of the meta-analyses can be seen in Table 42.

Table 42

Cognitive behavioural therapies versus TAU or control evidence summary.

Cognitive behavioural therapies versus other active intervention

Meta-analyses results revealed no significant difference between cognitive behavioural therapies and other therapies in maintaining abstinence both post-treatment and up to 15-month follow-up. A single study, however, did favour coping skills over counselling in the number of sober days at 12-month follow-up, and another single study favouring relapse prevention over psychotherapy at 15-month follow-up. However, these single outcomes do not reflect the meta-analyses results described above. In addition, cognitive behavioural therapies were found to be more effective at maintaining abstinence/light days when assessed up to 18-month follow-up (based on data by CONNORS2001). No significant difference was observed between groups in reducing heavy drinking episodes and the amount of alcohol consumed both post-treatment and up to 18-month follow-up. A single study outcome (ERIKSEN1986B) favoured coping skills over counselling in reducing the amount of alcohol consumed, but, again, this single study was not reflective of other analyses with similar variables.

The VEDEL2008 study assessed severity of relapse in their sample. The results indicated that other active intervention (namely CBT) was more effective than couples therapy (namely BCT) in reducing occasions in which participants lapsed (drank over six drinks on one occasion) or relapsed (drank more than six drinks most days of the week), but no significant difference was observed in the number of participants who relapsed on a regular basis (a few times a month). It must be noted that effect sizes were small and the results of a single study cannot be generalised.

No significant difference was observed between CBT and other active therapies in attrition rates.

The quality of this evidence is high, therefore further research is unlikely to change confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 43 and Table 44.

Table 43

Cognitive behavioural therapies versus other interventions evidence summary (1).

Table 44

Cognitive behavioural therapies versus other interventions evidence summary (2).

Comparing different formats of cognitive behavioural therapies

For maintaining abstinence, an individual assessment treatment programme was significantly more effective than a packaged CBT program when assessed post-treatment (moderate effect size, based on a single study). However, for the same comparison, no significant difference was observed between groups in reducing heavy drinking episodes. The addition of motivational enhancement to relapse prevention did not reduce the number of possible drinking days (at 6-month follow-up) and analyses favoured standard relapse prevention (moderate effect size). Furthermore, the addition of family therapy to coping skills did not show any significant benefit. Also, no significant difference in various drinking outcomes was observed between coping skills and other types of cognitive behavioural therapies (for example, cognitive behavioural mood-management training [CBMMT] when assessed at 6-month follow-up. No difference between CBT and coping skills were observed in the number of participants who had lapsed or relapsed at 6-month follow-up. No difference in attrition rates was observed between the various types of CBT.

More intensive coping skills was significantly better than standard coping skills at maintaining abstinent/light drinking at 12-month follow-up (moderate effect size) but this benefit was no longer significant at 18-month follow-up. Individual CBT was significantly more effective than group CBT in reducing the number of heavy drinkers at 15-month follow-up.

The quality of this evidence is moderate, therefore further research is likely to have an important impact on confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 45 and Table 46.

Table 45

Comparing different formats of CBT evidence summary.

Table 46

Comparing different formats of CBT evidence summary.

6.10.1. Definition

Behavioural interventions use behavioural theories of conditioning to help achieve abstinence from drinking by creating negative experiences/events in the presence of alcohol, and positive experiences/events in alcohol's absence. Behavioural therapies considered for review included cue exposure, behavioural self-control training, aversion therapy and contingency management. Variants of two therapies (cue exposure and behavioural self-control training) which were based on a similar theoretical understanding of the nature of alcohol misuse, were considered as a single entity for the purposes of the review. Contingency management, although a behavioural intervention, was analysed separately because it is based on the classic reinforcement model and has no alcohol specific formulation (see Section 6.11 for evidence review). Aversion therapy was excluded because it is no longer routinely used in alcohol-misuse treatment in the UK.

6.10.2. Clinical review protocol (behavioural therapies)

Information about the databases searched and the inclusion/exclusion criteria used for this section of the guideline can be found in Appendix 16d (further information about the search for health economic evidence can be found in Section 6.21). See Table 47, below, for a summary of the clinical review protocol for the review of behavioural therapies.

Table 47

Clinical review protocol for the review of behavioural therapies.

6.10.3. Studies considered for review

The review team conducted a systematic review of RCTs that assessed the beneficial or detrimental effects of behavioural therapies in the treatment of alcohol dependence or harmful alcohol use. See Table 48 for a summary of the study characteristics. It should be noted that some trials included in analyses were three- or four-arm trials.

Table 48

Summary of study characteristics for behavioural therapies.

To avoid double counting, the number of participants in treatment conditions used in more than one comparison was divided (by half in a three-arm trial, and by three in a four-arm trial).

Six RCT trials relating to clinical evidence met the eligibility criteria set by the GDG, providing data on 527 participants. All six studies were published in peer-reviewed journals between 1988 and 2006. A number of studies identified in the search were initially excluded because they were not relevant to this guideline. Studies were excluded because they did not meet the methodological criteria (see Chapter 3). When studies did meet basic methodological inclusion criteria, the main reasons for exclusion were not having alcohol-focused outcomes that could be used for analysis, and not meeting drinking quantity/diagnosis criteria, that is, participants were not drinking enough to be categorised as harmful or dependent drinkers or less than 80% of the sample meet criteria for alcohol dependence or harmful alcohol use. A list of excluded studies can be found in Appendix 16d.

6.10.4. Evidence summary

The GRADE profiles and associated forest plots for the comparisons can be found in Appendix 18c and Appendix 17c, respectively.

Behavioural therapies versus control/treatment as usual

The review evidence indicated that behavioural therapies were more effective than control in reducing the amount of alcohol consumed (SMD = −0.97, large effect size) and maintaining controlled drinking (SMD = −0.60, medium effect size) when assessed post-treatment. However, it must be noted that this was based on a single study.

No significant difference was observed between behavioural therapies and control in maintaining abstinence when assessed post-treatment. Furthermore, no significant difference could be found between behavioural therapies and control in the number of participants who lapsed or relapsed up to 6-month follow-up. In addition, there was no significant difference between behavioural therapies and control in attrition rates.

The quality of this evidence is moderate, therefore further research is likely to have an important impact on confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 49.

Table 49

Behavioural therapy versus TAU or control evidence summary.

Behavioural therapy versus other active intervention

The review evidence indicated that behavioural therapies were not as effective as other interventions (in this case, couples-based therapies) in maintaining abstinent/light drinking days up to 12-month follow-up. In addition, there was no significant difference between behavioural therapies and counselling in maintaining abstinence both post-treatment and up to 24-month follow-up.

No difference was observed between behavioural therapies and other active interventions (for example, CBT) in reducing the amount of alcohol consumed up to 24-month follow-up. However, one study (SITHARTHAN1997) showed a medium effect size favouring cue exposure over CBT in reducing drinks per occasion at 6-month follow-up.

Behavioural therapies were not as effective as other active interventions (namely couples therapies) in reducing heavy drinking days. Medium to high effects favouring couples therapy were found at all assessment points up to 12-month follow-up.

The review results revealed that other therapies (that is, CBT and counselling) had significantly less post-treatment attrition than behavioural therapies. However, no significant difference was observed between treatments at follow-up (3 to 24 months).

The quality of this evidence is moderate, therefore further research is likely to have an important impact on confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 50.

Table 50

Behavioural therapy versus other intervention evidence summary.

Comparing different formats of behavioural therapy

The clinical evidence indicates that there was no significant difference between cue exposure and BSCT in maintaining abstinence post-treatment or at 6-month follow-up. Furthermore, no significant difference was observed between cue exposure and emotional cue exposure in reducing the amount of alcohol consumed at 6- to 12-month follow-up. In line with this, no significant difference was observed between moderation-oriented cue exposure and behaviour self-control training in reducing alcohol consumption when assessed at 6-month follow-up.

No difference was observed between behavioural therapies in attrition both at post-treatment and 6-month follow-up.

The quality of this evidence is moderate, therefore further research is likely to have an important impact on confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 51.

Table 51

Comparing various formats of behavioural therapy evidence summary.

6.11.1. Definition

Contingency management provides a system of reinforcement designed to make continual alcohol use less attractive and abstinence more attractive. There are four main methods of providing incentives:

• Voucher-based reinforcement: people who misuse alcohol receive vouchers with various monetary values (usually increasing in value after successive periods of abstinence) for providing biological samples (usually urine) that are negative for alcohol. These vouchers are withheld when the biological sample indicates recent alcohol use. Once earned, vouchers are exchanged for goods or services that are compatible with an alcohol-free lifestyle.
• Prize-based reinforcement: This is more formally referred to as the ‘variable magnitude of reinforcement procedure’ (Prendergast et al., 2006). Participants receive draws, often from a number of slips of paper kept in a fishbowl, for providing a negative biological specimen. Provision of a specimen indicating recent alcohol use results in the withholding of draws. Each draw has a chance of winning a ‘prize’, and the value of which varies. Typically, about half the draws say ‘Good job!’. The other half results in the earning of a prize, which may range in value from £1 to £100 (Prendergast et al., 2006).
• Cash incentives: people who misuse alcohol receive cash (usually of a relatively low value, for example, £1.50 to £10) for performing the target behaviour, such as submitting a urine sample negative for alcohol or compliance with particular interventions. Cash incentives are withheld when the target behaviour is not performed.
• Clinic privileges: participants receive clinic privileges for performing the target behaviour, for example providing a negative biological sample. But these privileges are withheld when the target behaviour is not performed. An example of a clinic privilege is a take-home methadone dose (for example, Stitzer et al., 1992). This incentive is appropriate for drug treatment for substances such as heroin but is not applicable to alcohol treatment.

6.11.2. Clinical review protocol (contingency management)

Information about the databases searched and the inclusion/exclusion criteria used for this section of the guideline can be found in Chapter 3 (further information about the search for health economic evidence can be found in Section 6.21). See Table 52 below for a summary of the clinical review protocol for the review of contingency management).

Table 52

Clinical review protocol for the review of contingency management.

6.11.3. Studies considered for review

The review team conducted a systematic review of RCTs that assessed the beneficial or detrimental effects of contingency management in the treatment of alcohol dependence or harmful alcohol use. See Table 53 for a summary of the study characteristics.

Table 53

Summary of study characteristics for contingency management.

Three trials relating to clinical evidence met the eligibility criteria set by the GDG, providing data on 355 participants. All three studies were published in peer-reviewed journals between 2000 and 2007. A number of studies identified in the search were initially excluded because they were not relevant to this guideline. Studies were excluded because they did not meet methodological criteria (see Chapter 3). When studies did meet basic methodological inclusion criteria, the main reason for exclusion was that the participants in the study did not meet drinking quantity/diagnosis criteria (that is, participants were not drinking enough to be categorised as harmful or dependent drinkers, or less than 80% of the sample meet criteria for alcohol dependence or harmful alcohol use). Another reason was that the study was drug-focused or did not differentiate between drugs and alcohol. A list of excluded studies can be found in Appendix 16d.

6.11.4. Evidence summary

The GRADE profiles and associated forest plots for the comparisons can be found in Appendix 18c and Appendix 17c, respectively.

Contingency management versus control

The review evidence indicated that contingency management (with network support) was more effective at maintaining abstinence than control post-treatment (large effect size) and up to 15-month follow-up (medium effect size). However, no significant differences were observed between contingency management with network support and control for follow-up periods greater than 15 months. It should be noted that this analyses was based on the LITT2007 study only.

Contingency management (with network support) was more effective than control (low to medium effect size) at reducing drinking quantity when assessed at 6-, 9- and 21-month follow-up. However, no significant difference was found between treatment conditions post-treatment at 12-, 15-, 18-, 24- and 27-month follow-up.

No significant difference was observed between conditions in attrition post-treatment and at all follow-up points up to 27 months.

The quality of this evidence is moderate, therefore further research is likely to have an important impact on confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 54.

Table 54

Contingency management versus control evidence summary.

Contingency management versus treatment as usual (standard care)

The clinical review revealed no significant beneficial effect of adding contingency management to standard care in maintaining abstinence when assessed post-treatment. However, the addition of contingency management to standard care was beneficial in reducing the number of participants who relapsed to heavy drinking. Furthermore, the addition of contingency management to standard care was beneficial in reducing attrition rates.

The quality of this evidence is moderate, therefore further research is likely to have an important impact on confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 55.

Table 55

Contingency management versus standard care (TAU) evidence summary.

Contingency management versus other active interventions

The addition of contingency management to network support was not beneficial in maintaining abstinence both post-treatment and up to 9-month follow-up. However, network support without contingency management was more effective at maintaining abstinence at 12- to 24-month follow-up.

The quality of this evidence is moderate, therefore further research is likely to have an important impact on confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 56.

Table 56

Contingency management versus other intervention evidence summary.

6.12.1. Definition

Social network and environment-based therapies use the individual's social environment as a way to help achieve abstinence or controlled drinking. These therapies include SBNT and the community reinforcement approach.

6.12.2. Clinical review protocol (social network and environment-based therapies)

Information about the databases searched and the inclusion/exclusion criteria used for this section of the guideline can be found in Chapter 3 (further information about the search for health economic evidence can be found in Section 6.21). See Table 57 below for a summary of the clinical review protocol for the review of social network and environment-based therapies.

Table 57

Clinical review protocol for the review of social network and environment-based therapies.

6.12.3. Studies considered for review

The review team conducted a systematic review of RCTs that assessed the beneficial or detrimental effects of social network and environment-based therapies in the treatment of alcohol dependence or harmful alcohol use. See Table 58 for a summary of the study characteristics. It should be noted that some trials included in analyses were three- or four-arm trials. To avoid double counting, the number of participants in treatment conditions used in more than one comparison was divided (by half in a three-arm trial, and by three in a four-arm trial).

Table 58

Summary of study characteristics for social network and environment-based therapies.

Three trials relating to clinical evidence met the eligibility criteria set by the GDG, providing data on 1,058 participants. All three studies were published in peer-reviewed journals between 1999 and 2007. A number of studies identified in the search were initially excluded because they were not relevant to this guideline. Studies were excluded because they did not meet methodological criteria (see Chapter 3). When studies did meet basic methodological inclusion criteria, the main reason for exclusion was not having alcohol-focused outcomes that could be used for analysis. A list of excluded studies can be found in Appendix 16d.

6.12.4. Evidence summary

The GRADE profiles and associated forest plots for the comparisons can be found in Appendix 18c and Appendix 17c, respectively.

Social network and environment-based therapies versus control

The clinical evidence showed that social network and environment-based therapies were significantly better than control at maintaining abstinence (moderate effect size) when assessed post-treatment and at 6-, 9-, 12-, 15- and 24-month follow-up. However, no significant difference was observed at 18-, 21- and 27-month follow-up.

Social network and environment-based therapies were not significantly better than control in reducing drinking post-treatment or at 12-, 15-, 24- and 27-month follow-up. However, a significant benefit (low to moderate effect size) was observed for social network and environment-based therapies over control in reducing the quantity of alcohol consumed when assessed at 6-, 9-, 18- and 21-month follow-up.

No significant difference was observed between treatment conditions in attrition either post-treatment or at all follow-up points. It must be noted that the comparison between social network and environment-based therapies versus control was based on a single study.

The quality of this evidence is moderate, therefore further research is likely to have an important impact on confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 59.

Table 59

Social network and environment-based therapies versus control evidence summary.

Social network and environment-based therapies versus other active intervention

The clinical evidence did not reveal any significant difference between social network and environment-based therapies and other active interventions in maintaining abstinence, reducing the quantity of alcohol consumed, reducing the number of drinking days and attrition.

The quality of this evidence is moderate therefore further research is likely to have an important impact on confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 60.

Table 60

Social network and environment-based therapies versus other intervention evidence summary.

6.13.1. Definition

The content and definition of couples therapy can vary and reflect different approaches, for example cognitive behavioural or psychodynamic. Couples-based interventions (including behavioural couples therapy [BCT]) involve the spouse or partner expressing active support for the person who misuses alcohol in reducing alcohol use, including support via the use of behavioural contracts. Couples are helped to improve their relationship through more effective communication skills, and encouraged to increase positive behavioural exchanges through acknowledgement of pleasing behaviours and engagement in shared recreational activities (Fals-Stewart et al., 2005). Standard BCT is manual-based and structured (Fals-Stewart et al., 2004), and combines cognitive-behavioural treatment strategies with methods that address relationship issues arising from alcohol misuse as well as more general relationship problems with the aim of reducing distress.

6.13.2. Clinical review protocol (couples therapy)

Information about the databases searched and the inclusion/exclusion criteria used for this section of the guideline can be found in Chapter 3 (further information about the search for health economic evidence can be found in Section 6.21). See Table 61 below for a summary of the clinical review protocol for the review of couples therapy.

Table 61

Clinical review protocol for the review of couples therapy.

6.13.3. Studies considered for review

The review team conducted a systematic review of RCTs that assessed the beneficial or detrimental effects of couples therapies in the treatment of alcohol dependence or harmful alcohol use. See Table 62 for a summary of the study characteristics. It should be noted that some trials included in analyses were three- or four-arm trials. To avoid double counting, the number of participants in treatment conditions used in more than one comparison was divided (by half in a three-arm trial, and by three in a four-arm trial).

Table 62

Summary of study characteristics for couples therapy.

Eight trials relating to clinical evidence met the eligibility criteria set by the GDG, providing data on 602 participants. All eight studies were published in peer-reviewed journals between 1988 and 2009. A number of studies identified in the search were initially excluded because they were not relevant to this guideline. Studies were excluded because they did not meet methodological criteria (see Chapter 3). When studies did meet basic methodological inclusion criteria, the main reason for exclusion was not having alcohol-focused outcomes that could be used for analysis. Other reasons were not meeting drinking quantity/diagnosis criteria, that is, participants were not drinking enough to be categorised as harmful or dependent drinkers, or less than 80% of the sample meet criteria for alcohol dependence or harmful alcohol use, the study was outside the scope of this guideline, or the study was drug-focused or did not differentiate between drugs and alcohol. A list of excluded studies can be found in Appendix 16d.

6.13.4. Evidence summary

The GRADE profiles and associated forest plots for the comparisons can be found in Appendix 18c and Appendix 17c, respectively.

Couples therapy versus other active intervention

Not significant difference was observed between couples therapy (all types) and other active interventions in maintaining abstinence at post-treatment and 2-month follow-up assessment. However, over longer periods, couples therapy was significantly more effective than other therapies in maintaining abstinence and/or light drinking (moderate effect size) when assessed up to 12-month follow-up. This difference was not observed in follow-up periods longer than 12 months. An additional randomised study (MCCRADY2009) could not be included in these analyses as no extractable data was provided. The study reported that BCT was more effective than individual coping-skills treatment in maintaining abstinence and reducing heavy drinking days.

Couples therapy was significantly more effective than other active interventions in reducing heavy drinking episodes when assessed up to 12-month follow-up. However, there was no difference between couples therapy and other active interventions post-treatment.

The VEDEL2008 study assessed severity of relapse in their sample. The results indicated that other active intervention (namely CBT) was more effective than couples therapy (namely BCT) in reducing occasions in which participants lapsed (drank over six drinks on one occasion) or relapsed (drank more than six drinks most days of the week), but no significant difference was observed in the number of participants who relapsed on a regular basis (a few times a month). It must be noted that effect sizes were small and from a single study.

No difference in attrition rates was observed between groups post-treatment and at 3-month follow-up. Couples therapy had less attrition than other therapies at 6-month follow-up (large effect size), and other therapies had less attrition than couples therapy at 12-month follow-up (large effect size).

The quality of this evidence is moderate, therefore further research is likely to have an important impact on the confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 63.

Table 63

Couples therapy versus other intervention evidence summary.

Behavioural couples therapy versus other couples therapy

No significant difference was observed between BCT and other forms of couples therapy in maintaining abstinence when assessed post-treatment and up to 24-month follow-up. Similarly, no difference between these groups was observed in reducing heavy drinking and attrition rates post-treatment, and up to 12-month follow-up.

The quality of this evidence is moderate, therefore further research is likely to have an important impact on the confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 64.

Table 64

Behavioural couples therapy versus other couples therapy evidence summary.

Intensive versus standard couples therapy

At 1-month follow-up, brief couples therapy was more effective than more intensive couples therapy in maintaining abstinence (moderate effect size). However, this difference was not maintained up to 18-month follow-up. Furthermore, no significant benefit of more intensive couples therapy over brief couples therapy in reducing heavy drinking was observed up to 18-month follow-up. Those who received more intensive couples therapy were more likely to be retained for follow-up assessment at 12 months than brief couples therapy (small effect size).

The quality of this evidence is moderate, therefore further research is likely to have an important impact on the confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 65.

Table 65

Intensive couples therapy versus brief couples therapy evidence summary.

Parental skills and behavioural couples therapy versus behavioural couples therapy alone

The addition of parental skills training to BCT did not significant improve abstinence rates both post-treatment and up to 12-month follow-up.

The quality of this evidence is moderate, therefore further research is likely to have an important impact on the confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 66.

Table 66

Parental skills and behavioural couples therapy versus behavioural couples therapy alone evidence summary.

6.14.1. Definition

The British Association for Counselling and Psychotherapy defines counselling as ‘a systematic process which gives individuals an opportunity to explore, discover and clarify ways of living more resourcefully, with a greater sense of well-being’ (British Association of Counselling, 1992). This definition, which has been used in other NICE guidelines, was adopted for this review, but in the included studies counselling for alcohol treatment was not often well-defined or manual-based making decisions about inclusion difficult; where there was uncertainty this was resolved in discussion with the GDG.

6.14.2. Clinical review protocol (counselling)

Information about the databases searched and the inclusion/exclusion criteria used for this section of the guideline can be found in Chapter 3 (further information about the search for health economic evidence can be found in Section 6.21). See Table 67, below, for a summary of the clinical review protocol for the review of counselling.

Table 67

Clinical review protocol for the review of counselling.

6.14.3. Studies considered for review

The review team conducted a systematic review of RCTs that assessed the beneficial or detrimental effects of counselling in the treatment of alcohol dependence or harmful alcohol use. See Table 68 for a summary of the study characteristics. It should be noted that some trials included in analyses were three- or four-arm trials. To avoid double counting, the number of participants in treatment conditions used in more than one comparison was divided (by half in a three-arm trial, and by three in a four-arm trial).

Table 68

Summary of study characteristics for counselling.

Five trials relating to clinical evidence met the eligibility criteria set by the GDG, providing data on 630 participants. All five studies were published in peer-reviewed journals between 1986 and 2003. A number of studies identified in the search were initially excluded because they were not relevant to this guideline. Counselling studies were mainly excluded for not being randomised trials. When studies did meet basic methodological inclusion criteria, the main reason for exclusion were that treatment was opportunistic as opposed to planned, the study was not directly relevant to the review questions, or no relevant alcohol-focused outcomes were available. A list of excluded studies can be found in Appendix 16d.

6.14.4. Evidence summary

The GRADE profiles and associated forest plots for the comparisons can be found in Appendix 18c and Appendix 17c, respectively.

Counselling versus Control

Based on the SELLMAN2001 study, no significant difference was observed between treatment groups, hence the clinical evidence does not support the benefits of counselling over control in maintaining abstinence or reducing heavy drinking.

The quality of this evidence is moderate, therefore further research is likely to have an important impact on confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 69.

Table 69

Counselling versus control evidence summary.

Counselling versus other active intervention

In maintaining abstinence, no significant difference was observed between counselling and other therapies when assessed up to 6-month follow-up. However, bar the 6-month follow-up, these results are based on a single study (OFARRELL1992) whereas in the analyses assessing couples therapies versus other active therapies, more studies were included in the analyses for this outcome. Other therapies (namely couples therapies and coping skills) showed significant benefits over counselling in maintaining abstinence at longer follow-up periods of up to 18 months.

Overall, no significant difference was observed between counselling and other therapies up to 18-month follow-up in time to first drink (lapse), time to first heavy drink (relapse) and reducing heavy drinking episodes. These analyses were based on data from a single study (LITT2003). However, other therapies (coping skills) were more effective than counselling in reducing amount of alcohol consumed when assessed at 12-month follow-up. Again, this result was based on a single study (ERIKSEN1986B) limiting the ability to generalise the findings.

Lastly, no significant difference was observed between counselling and other therapies in attrition rates.

The quality of this evidence is moderate, therefore further research is likely to have an important impact on confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 70.

Table 70

Counselling versus other intervention evidence summary.

6.15.1. Definition

Short-term psychodynamic therapy is a derived from a psychodynamic/psychoanalytic model in which: (a) therapist and patient explore and gain insight into conflicts, and how these are represented in current situations and relationships, including the therapeutic relationship; (b) service users are given an opportunity to explore feelings, and conscious and unconscious conflicts originating in the past, with the technical focus on interpreting and working through conflicts; and (c) therapy is non-directive and service users are not taught specific skills such as thought monitoring, re-evaluation or problem solving. Treatment typically consists of 16 to 30 sessions (Leichsenring et al., 2004), but there are interventions that offer more or less than this range.

6.15.2. Clinical review protocol (short-term psychodynamic therapy)

Information about the databases searched and the inclusion/exclusion criteria used for this section of the guideline can be found in Chapter 3 (further information about the search for health economic evidence can be found in Section 6.21). See Table 71 for a summary of the clinical review protocol for the review of short-term psychodynamic therapy).

Table 71

Clinical review protocol for the review of short-term psychodynamic therapy.

6.15.3. Studies considered for review

The review team conducted a systematic review of RCTs that assessed the beneficial or detrimental effects of psychodynamic therapies in the treatment of alcohol dependence or harmful alcohol use. See Table 72 for a summary of the study characteristics.

Table 72

Summary of study characteristics for short-term psychodynamic therapy.

One trial relating to clinical evidence met the eligibility criteria set by the GDG, providing data on 49 participants. The study was published in a peer-reviewed journal in 1998. A number of studies identified in the search were initially excluded because they were not relevant to this guideline. Studies were further excluded because they did not meet methodological criteria (see Chapter 3). When studies did meet basic methodological inclusion criteria, the main reasons for exclusion were that the study was not directly relevant to the review questions, or no relevant alcohol-focused outcomes were available. A list of excluded studies can be found in Appendix 16d.

6.15.4. Evidence summary

The GRADE profiles and associated forest plots for the comparisons can be found in Appendix 18c and Appendix 17c, respectively.

Short-term psychodynamic therapy versus other active intervention

At 15-month follow-up, short-term psychodynamic therapy was significantly more effective than other therapies (in this case, cognitive behavioural relapse prevention) in maintaining abstinence, although the effect size was moderate. However, no significant difference was observed between short-term psychodynamic therapy and other therapies in reducing the quantity of alcohol consumed, heavy drinking rate or attrition. It must be noted that this analysis was based on a single study.

The quality of this evidence is moderate, therefore further research is likely to have an important impact on the confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 73.

Table 73

Short-term psychodynamic therapy versus other intervention evidence summary.

6.16.1. Definition

Multi-modal treatment for alcohol misuse involves a combination of a number of interventions that have been developed and evaluated as stand-alone interventions for alcohol misuse. Components of a multi-modal treatment could include motivational aspects (such as MET), TSF, AA or self-help group participation, group counselling, CBT-based relapse-prevention training and psychoeducational sessions. The intention is that by combining a number of effective interventions the combined treatment will be greater than any one individual treatment.

6.16.2. Clinical review protocol (multi-modal treatment)

Information about the databases searched and the inclusion/exclusion criteria used for this section of the guideline can be found in Chapter 3 (further information about the search for health economic evidence can be found in Section 6.21).

Table 74

Clinical review protocol for the review of multi-modal treatment.

6.16.3. Studies considered for review

The review team conducted a systematic review of RCTs that assessed the beneficial or detrimental effects of multi-modal therapies in the treatment of alcohol dependence or harmful alcohol use. See Table 75 for a summary of the study characteristics.

Table 75

Summary of study characteristics for multi-modal treatment.

Two trials relating to clinical evidence met the eligibility criteria set by the GDG, providing data on 427 participants. Both studies were published in peer-reviewed journals between 2002 and 2003. A number of studies identified in the search were initially excluded because they were not relevant to this guideline. Studies were excluded because they did not meet methodological criteria (see Chapter 3). When studies did meet basic methodological inclusion criteria, the main reason for exclusion was that no relevant alcohol-focused outcomes were available. A list of excluded studies can be found in Appendix 16d.

6.16.4. Evidence summary

The GRADE profiles and associated forest plots for the comparisons can be found in Appendix 18c and Appendix 17c, respectively.

Multi-modal versus other active intervention

A small effect was observed favouring other therapies (that is, psychoeducational) over multi-modal treatment in maintaining abstinence when assessed post-treatment. In addition, other therapies (that is, counselling) were significantly better than multi-modal treatment in reducing the number of participants who had lapsed (small effect size). However, this was not the case at 12-month follow-up because no difference between groups was observed. Furthermore, no difference was observed between multi-modal treatment and other therapies in reducing the number of days drinking, the quantity of alcohol consumed and attrition up to 12-month follow-up.

The quality of this evidence is low, therefore further research is very likely to have an important impact on the confidence in the estimate of the effect and is likely to change the estimate. An evidence summary of the results of the meta-analyses can be seen in Table 76.

Table 76

Multimodal intervention versus other intervention evidence summary.

6.17.1. Definition

A self-help intervention is where a healthcare professional (or para-professional) would facilitate the use of the self-help material by introducing, monitoring and reviewing the outcome of such treatment. The intervention is limited in nature–usually no more than three to five sessions, some of which may be delivered by telephone. Self-administered interventions are designed to modify drinking behaviour and make use of a range of books, web pages, CD-ROMs or a self-help manual that is based on an evidence-based intervention and designed specifically for the purpose. An example is guided self-change (Sobell & Sobell, 1993). This treatment is manual-based and uses the principles of CBT and MET. The patient has an initial assessment followed by four treatment sessions and two follow-up telephone calls.

6.17.2. Clinical review protocol (self-help-based treatment)

Information about the databases searched and the inclusion/exclusion criteria used for this section of the guideline can be found in Chapter 3 (further information about the search for health economic evidence can be found in Section 6.21). See Table 77 for a summary of the clinical review protocol for the review of self-help-based treatment.

Table 77

Clinical review protocol for the review of self-help-based treatment.

6.17.3. Studies considered for review

The review team conducted a systematic review of RCTs that assessed the beneficial or detrimental effects of self-help-based treatment in the treatment of alcohol dependence or harmful alcohol use. See Table 78 for a summary of the study characteristics.

Table 78

Summary of study characteristics for self-help-based treatment.

One trial relating to clinical evidence met the eligibility criteria set by the GDG, providing data on 93 participants. The included study was published in a peer-reviewed journal in 2002. A number of studies identified in the search were initially excluded because they were not relevant to this guideline. Studies were excluded because they did not meet methodological criteria (see Chapter 3). A particular problem for self-help-based treatments is that they usually fall under the grouping of ‘brief interventions’. Therefore, the main reasons for exclusions were the population assessed were hazardous drinkers (outside the scope of this guideline), the population were not treatment seeking, or no relevant alcohol-focused outcomes were available. A list of excluded studies can be found in Appendix 16d.

6.17.4. Evidence summary

The GRADE profiles and associated forest plots for the comparisons can be found in Appendix 18c and Appendix 17c, respectively.

Guided self-help-based treatment (guided) versus non-guided self-help-based treatment

Guided self-help was significantly more effective than non-guided self-help in reducing the quantity of drinks consumed per week when assessed at 9-month follow-up. However, no significant difference was observed between groups for the same variable at 23-month follow-up, or for the number of DDD (at 9- and 23-month follow-up) or attrition at 23-month follow-up.

The quality of this evidence is moderate, therefore further research is likely to have an important impact on the confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 79.

Table 79

Comparing different formats of self-help-based treatment evidence summary.

6.18.1. Definition

A psychoeducational intervention involves an interaction between an information provider and service user. It has the primary aim of offering information about the condition, as well as providing support and management strategies. Psychoeducational intervention for alcohol misuse involves the use of education videos, literature and lectures that highlight the health and lifestyle risks of excessive alcohol consumption. It is not usually used as a formal method of treatment, but an adjunct to conventional treatment methods. Psychoeducational attention control treatment (PACT) is a form of manual-based psychoeducational therapy developed by Fals-Stewart and Klostermann (2004) and used in some alcohol treatment trials.

6.18.2. Clinical review protocol (psychoeducational interventions)

Information about the databases searched and the inclusion/exclusion criteria used for this section of the guideline can be found in Chapter 3 (further information about the search for health economic evidence can be found in Section 6.21). See Table 80 for a summary of the clinical review protocol for the review of psychoeducational interventions.

Table 80

Clinical review protocol for the review of psychoeducational interventions.

6.18.3. Studies considered for review

The review team conducted a systematic review of RCTs that assessed the beneficial or detrimental effects of behavioural therapies in the treatment of alcohol dependence or harmful alcohol use. See Table 50 (above, in Section 6.11) for a summary of the study characteristics. It should be noted that some trials included in analyses were three- or four-arm trials. To avoid double counting, the number of participants in treatment conditions used in more than one comparison was divided (by half in a three-arm trial, and by three in a four-arm trial).

Five trials relating to clinical evidence met the eligibility criteria set by the GDG, providing data on 1,312 participants. All five studies were published in peer-reviewed journals between 2001 and 2006. A number of studies identified in the search were initially excluded because they were not relevant to this guideline. Studies were excluded because they did not meet methodological criteria (see Chapter 3). When studies did meet basic methodological inclusion criteria, the main reason for exclusion was not meeting drinking quantity/diagnosis criteria, that is, participants were not drinking enough to be categorised as harmful or dependent or less than 80% of the sample meet criteria for alcohol dependence or harmful alcohol use. A list of excluded studies can be found in Appendix 16d.

Psychoeducational interventions versus other active intervention

All five included trials assessed psychoeducational interventions versus another active intervention. CONNORS2001 was complex in design and investigated a psychoeducational intervention plus alcohol-focused coping skills versus life skills plus alcohol-focused coping skills. Additionally, the study investigated the difference between low- and high-intensity treatment of these conditions. The results of the 30-month follow-up were obtained from Walitzer and Connors (2007). DAVIS2002 assessed a psychoeducational intervention versus standard multi-modal treatment. FALSSTEWART2005 investigated a psychoeducational intervention (used as an attentional control) versus BCT (plus group counselling), brief relationship therapy (plus group counselling) and individually-based TSF (plus group counselling). FALSSTEWART2006 investigated a psychoeducational intervention (as an attentional control) versus BCT (plus individually-based TSF) as well as individually-based TSF alone. SOBELL2002 investigated psychoeducational (bibliotherapy/drinking guidelines) versus motivational enhancement/personalised feedback. See Table 81 for a summary of the study characteristics.

Table 81

Summary of study characteristics for psychoeducational intervention.

6.18.4. Evidence summary

The GRADE profiles and associated forest plots for the comparisons can be found in Appendix 18c and Appendix 17c, respectively.

Psychoeducational versus other active intervention

The clinical findings for this comparison are mixed whether in favour of other active intervention over a psychoeducational intervention or finding no clinically significant difference between psychoeducational and other interventions. Other interventions were significantly better than psychoeducational interventions in increasing length of sobriety (post-treatment), and the percentage of abstinent/light drink days at 6- and 12-month follow-up.

No significant difference was observed between a psychoeducational intervention and other active interventions in attrition rates and other drinking-related variables.

The quality of this evidence is moderate, therefore further research is likely to have an important impact on the confidence in the estimate of the effect. An evidence summary of the results of the meta-analyses can be seen in Table 82.

Table 82

Psychoeducational intervention versus other intervention evidence summary.

6.19.1. Definition

Mindfulness meditation is rooted in the principles of Buddhism and is characterised by having a non-judgemental approach to experiences that result in the practitioner acting reflectively rather than impulsively on these experiences (Chiesa, 2010). Mindfulness meditation has a goal of developing a non-judgemental attitude and relationship to thoughts, feelings and actions as they are experienced by the practitioner, and not necessarily to change the content of thoughts as in CBT for example (Teasdale et al., 1995).

Mindfulness-based meditation has been suggested as a method of improving physical and mental health (for a review, see Allen et al., 2006). However, the quality of this research is generally poor, not focused on alcohol as the substance of misuse, and few in number.

6.19.2. Clinical review protocol (mindfulness meditation)

In the current review, the role of meditation in maintaining abstinence and drinking reduction was investigated. Its application to other aspects usually associated with alternative therapies in this topic area (such as craving and withdrawal symptoms) was beyond the scope of this guideline and hence was not investigated. Information about the databases searched and the inclusion/exclusion criteria used for this section of the guideline can be found in Chapter 3. See Table 83 for a summary of the clinical review protocol for the review of meditation.

Table 83

Clinical review protocol for the review of meditation.

6.19.3. Studies considered for review

The review team conducted a systematic search of RCTs and systematic reviews that assessed the beneficial or detrimental effects of meditation in the treatment of alcohol dependence or harmful alcohol use. Following the literature search, there was an insufficient number of studies remaining to perform an unbiased and comprehensive meta-analysis of meditation for the treatment of alcohol misuse. Therefore, the GDG consensus was that a narrative summary of these studies would be conducted and observational studies would be included in the review. See Table 84 for a summary of the study characteristics.

Table 84

Summary of study characteristics for mindfulness meditation.

Two trials (BOWEN2006³⁷; ZGIERSKA2008) providing data on 320 participants were identified by the search. Both studies were published in peer-reviewed journals between 2006 and 2008. To the GDG's knowledge, no other studies that evaluated meditation for a population with alcohol misuse with alcohol-focused outcomes have been published.

6.19.4. Evidence summary

Bowen and colleagues (2006) investigated the effectiveness of mindfulness meditation on substance-use outcomes in an incarcerated population. The study compared mindfulness meditation with treatment as usual (a chemical dependency programme and psychoeducational intervention). The authors reported that mindfulness meditation was significantly more effective than treatment as usual in the amount of alcohol consumed at 3-month follow-up (p <0.005). However, adherence to the therapy was not assessed and therefore the authors were unclear whether participants correctly followed the principles of mindfulness meditation. Furthermore, the level of alcohol dependence in this sample was unclear.

In a feasibility pilot prospective case series study, Zgierska and colleagues (2008) evaluated the efficacy of mindfulness meditation in increasing abstinence and reducing the quantity of alcohol consumed. Alcohol-dependent participants who had recently completed an intensive outpatient treatment programme were recruited. The study found that participants reported significantly fewer heavy drinking days at 4-, 8- and 12-week follow-up (all p <0.005), but not at 16-week follow-up. Furthermore, participants were drinking substantially less when assessed at 4- and 8-week follow-up (p <0.005), but no significant difference was observed at 12- and 16-week follow-up. No significant difference over time was observed in increasing PDA. It must be noted, however, that meditation in this study was not used as an active intervention but as a post-treatment intervention. Furthermore, the sample size was small and the study had no control group.

These studies reported a significant effect of mindfulness meditation on alcohol consumption. Overall, there is limited and poor-quality evidence that does not support the use of mindfulness-based meditation for treating alcohol dependence and harmful alcohol use.

6.21.1. Review overview

The literature search identified six studies that assessed the cost effectiveness of psychological interventions for the treatment of alcohol dependence or harmful alcohol use (Alwyn et al., 2004; Fals-Stewart et al., 2005; Holder et al., 2000; Mortimer & Segal 2005; Slattery et al., 2003; UKATT Research Team, 2005). Full references, characteristics and results of all studies included in the economic review are presented in the form of evidence tables in Appendix 19.

The study by Alwyn and colleagues (2004) considered the cost effectiveness of adding a psychological intervention to a conventional home detoxification programme for the treatment of ‘problem drinkers’. The home detoxification programme comprised five home visits of 30 minutes duration delivered by CPNs. The study population consisted of 91 heavy drinkers in the UK who fulfilled inclusion criteria for home detoxification. A number of outcome measures were assessed in the study including: number of DDD; total number of days abstinent; total number of alcohol units consumed; abstinence or moderate drinking; and severity of dependence. The NNT to produce one extra non-drinker was also calculated. An NHS perspective was used for the economic analysis. Resource-use data included inpatient days, outpatient care (including CPN visits) and medications. Because clinical outcomes were left disaggregated and no summary outcome measure was used in the economic analysis, a cost-consequences analysis was used.

The authors made no formal attempt to compare the total costs of a psychological intervention in addition to home detoxification with home detoxification alone. Instead the authors calculated total costs per person of inpatient treatment (£2,186 to £3,901), outpatient treatment (£581 to £768) and home detoxification plus psychological intervention (£231). Therefore, the extra cost of a psychological intervention programme was substantially lower than the cost of inpatient treatment and outpatient visits. In terms of clinical outcomes, significantly better results were observed in people treated with home detoxification plus a psychological intervention. The authors concluded that, due to the low NNT to obtain an extra non-drinker, it is likely that the implementation of a psychological intervention would lead to cost savings to the NHS. Although the results of this study are highly relevant to the UK context, there are a number of methodological limitations. Firstly, no attempt was made to combine costs and effectiveness with an array of effectiveness measures used in the study. The measures of effectiveness used are of limited usefulness to policy makers when assessing the comparative cost effectiveness of healthcare interventions. The clinical-effectiveness study compared psychological intervention and home detoxification with home detoxification alone. However, in the cost-analysis, home detoxification was compared with other detoxification programmes, such as inpatient and outpatient programmes. Therefore, the study did not directly assess the cost-effectiveness of adding psychological intervention to home detoxification.

The study by Fals-Stewart and colleagues (2005) considered the cost-effectiveness of brief relationship therapy (BRT) compared with standard BCT (S-BCT), individual-based treatment (IBT) and PACT for alcohol-dependent males and their non-substance-misusing female partners. BRT, IBT and PACT consisted of 18 therapy sessions over 12 weeks whilst S-BCT consisted of 24 sessions over 12 weeks with participants randomised to one of the four groups and followed up for 12 months. A societal perspective was taken for the analysis with costs including those associated with the four treatment programmes and costs of participants' and their partners' travel time. The primary measure of effectiveness for the economic analysis was the change in percentage days of heavy drinking from baseline to 12 months. Rather than calculating ICERs, the authors calculated mean change in PDHD over 12 months divided by mean cost of treatment delivery (in US$100 units), with higher ratios indicating greater cost effectiveness. Overall, BRT had the highest mean ratio (4.61) of the four treatment programmes considered, suggesting this was the most cost-effective treatment. The findings of this study have limited applicability to this guideline as it is based within the US health system, outcomes were not expressed as QALYs and a societal perspective was taken for the cost analysis (both outside the NICE reference case). Furthermore, no formal attempt at an incremental analysis, in terms of differences in costs and outcomes, was attempted by the authors.

The study by Holder and colleagues (2000) compared the healthcare costs of three treatment modalities (12-session CBT, 4-session MET and 12-session TSF) over 3 years' follow-up. The study participants were a sample (65%) of individuals with alcohol dependency symptoms taken from the US Project MATCH study (Project MATCH Research Group, 1998). The perspective of the cost analysis was from US healthcare providers. Resource-use data included the three treatments and any subsequent inpatient or outpatient care over 3 years. The authors calculated mean monthly costs for the three treatments rather than total costs over 3 years and no incremental or statistical analyses were presented. Overall, mean monthly costs were US$186 for CBT, US$176 for MET and US$225 for TSF, suggesting that MET had the largest potential healthcare savings over 3 years. The major limitations of this analysis were the lack of descriptive detail on the resource use and costs considered whilst no incremental analysis was presented. The findings have limited applicability to this guideline as it was based on the US healthcare system and no formal attempt was made by the authors to combine cost and clinical-outcomes data, which were collected in the study and reported elsewhere (Project MATCH Research Group, 1998).

The study by Mortimer and Segal (2005) conducted separate, mutually exclusive, model-based economic analyses of interventions for ‘problem drinking’ and alcohol dependence. A lifetime horizon was used for all of the analyses considered. The first analysis compared three brief motivational interventions with different levels of intensity (simple was 5 minutes, brief was 20 minutes and extended was four sessions of 120 to 150 minutes) with no active treatment in a population of heavy drinkers within the Australian healthcare setting. The outcome measure used in the analysis was QALYs calculated from disability weights derived from a single published source (Stouthard et al., 1997). Clinical effectiveness data were taken from published studies evaluating interventions that were targeting heavy drinkers at lower severity levels. These data were used to estimate how patients would progress between specific drinking states (problem, moderate or dependent) within the model. The authors did not specify the resource use and cost components included in the model, although a health service perspective was adopted for the analysis. The results of the analysis suggested that brief motivational interventions were cost effective compared with no active treatment. The ICERs ranged from under AUS$82 (£61) per QALY for the simple intervention to under AUS$282 (£179) per QALY for the extended intervention.

The second analysis compared psychotherapies for mild to moderate alcohol dependence. The comparators were moderation-oriented cue exposure (MOCE) versus BSCT and MET or non-directive reflective listening versus no further counselling after initial assessment, also within the Australian healthcare setting. Again, the outcome measure used in the analysis was the QALY calculated from disability weights derived from a single published source (Stouthard et al., 1997). Clinical effectiveness data were taken from published studies evaluating interventions for mild to severe dependence. These data were used to estimate how patients would progress between specific drinking states (problem, moderate or dependent) within the model. No resource use and cost components were specified within the article. The results of the analysis suggested that MOCE was cost effective in comparison with BSCT, resulting in an ICER of AUS$2,145 (£1,589) per QALY. Non-directive reflective listening was dominated by no further counselling after initial assessment, resulting in higher costs but lower QALYs. However, the results of the analysis suggested that MET was cost effective compared with no further counselling after initial assessment, resulting in an ICER of AUS$3366 (£2,493) per QALY.

There are several limitations of the results of the study by Mortimer and Segal (2005) that reduce their applicability to any UK-based recommendations. In the second analysis of interventions for mild to moderate alcohol dependence, a common baseline comparator was not used in the analyses of MOCE, MET and non-directive reflective listening, limiting their comparability in terms of cost effectiveness. Ideally, indirect comparisons of the three interventions would have provided additional information about their relative effectiveness. Little explanation was given in the article as to how the clinical effectiveness data, which were taken from various sources, were used to inform the health states used in the economic models. The article did not specify the resource use and costs that were included in the analyses although a health perspective was used. The analyses all used QALYs as the primary outcome measure, which allows for comparison across interventions, although again there was insufficient description of the utility weights that were applied to the health states within the model.

The study by Slattery and colleagues (2003) developed an economic model to assess the cost effectiveness of four psychological interventions in comparison with standard care within the Scottish health service: coping/social skills training; BSCT; MET and marital/family therapy. The populations examined were 45-year-old men and women with a diagnosis of alcohol dependence. The outcome measures used in the economic model were the number of people who have abstained and number of deaths averted. The clinical effectiveness data were based on a methodologically diverse selection of trials which were not described within the study. Most studies included a treatment arm in which the intervention was thought likely to have little or no effect and this was used as the comparator arm when available. Resource use involved in the delivery of psychological interventions was estimated from expert clinical opinion and included the number and duration of sessions, staff and educational materials. Unit costs were taken from Scottish health service estimates. Other healthcare costs included in the model were those associated with alcohol-related disease endpoints such as stroke, cancer, cirrhosis and alcohol-related psychoses. Costs were applied according to inpatient length of stay taken from Scottish medical records.

For each intervention, the costs of psychological treatment and any disease endpoints for a hypothetical cohort of 1000 patients were compared with standard care over a 20-year time horizon, to determine any net healthcare cost savings. All four therapies demonstrated net savings ranging from £274,008 (coping/social skills training) to £80,452 (BSCT) in comparison with standard care. All four interventions resulted in lower costs per additional abstinent person and lower costs per death averted in comparison with standard care. Whilst the results of the study, based on a hypothetical cohort of patients within the Scottish health service, may be applicable to a UK setting, there are several problematic methodological issues with the study. First, the sources of the effectiveness data used in the model were not explicitly described by the authors, who suggested that the data were taken from a methodologically diverse selection of trials, thus suggesting a high level of heterogeneity. Second, no attempt was made to translate intermediate clinical endpoints such as abstinence rates into QALYs, which are useful to decision makers when assessing the comparative cost effectiveness of healthcare interventions.

The UKATT study (2005) evaluated the cost effectiveness of MET versus SBNT amongst a population comprising people who would normally seek treatment for alcohol misuse at UK treatment sites. The outcome measure used in the economic analysis was the QALY, which was estimated by using the EQ-5D questionnaire completed by patients at baseline, 3 and 12 months. The primary measures of clinical effectiveness were changes in alcohol consumption, alcohol dependence and alcohol-related problems over the 12-month period. A societal perspective was taken for the analysis. Resource-use data that were collected during the study included training and supervision, and materials related to treatment, hospitalisation, outpatient visits, GP and CPN visits, rehabilitation and consultation in alcohol agencies, social service contacts and court attendances. Unit-cost estimates were derived from a variety of UK published sources.

At 12 months, the total mean costs were higher in the MET group, resulting in a mean difference of £206 per patient (95% CI, −£454 to £818) versus SBNT. After adjusting for baseline differences, the MET group achieved slightly higher QALYs than SBNT, resulting in a mean difference of 0.0113 QALYs (95% CI, −0.0532 to 0.0235). Combining costs and QALYs, the MET group had an ICER of £18,230 in comparison with SBNT. CEACs showed that, at a cost-effectiveness threshold of £30,000 per QALY, MET had a 57.6% probability of being more cost effective than SBNT. The results of the study are applicable to a UK setting and the outcome measure used enables comparison across healthcare interventions. However, as the authors note, the analysis had a short time horizon and the longer-term effects of a reduction in drinking were not taken into consideration.

6.21.2. Health economic summary

The systematic search of the health economic literature did not identify evidence on the cost effectiveness of all of the psychological interventions considered in this guideline. Three of the studies identified were UK-based (Alwyn et al., 2004; Slattery et al., 2003; UKATT Research Team, 2005), two were US-based (Fals-Stewart et al., 2005; Holder et al., 2000) and one was Australian (Mortimer & Segal, 2005). The study by Alwyn and colleagues (2004) suggested that adding a psychological intervention to a home detoxification programme may offer the NHS cost savings in ‘problem drinkers’. The study by Slattery and colleagues (2003) showed that four psychological interventions, including coping/social skills training, BSCT, MET and marital/family therapy, offered significant healthcare cost savings compared with standard care for alcohol-dependent patients. The UKATT Research Team (2005) suggested that MET was cost effective in people who misuse alcohol, at current UK thresholds, in comparison with SBNT (but note that it was not identified as a clinically effective intervention in this guideline). Fals-Stewart and colleagues (2005) concluded that brief relationship therapy was significantly more cost effective compared with standard BCT, IBT and psychoeducational control treatment. Holder and colleagues (2000) suggested that MET offered the largest potential healthcare cost savings over 3 years when compared with CBT or TSF. Mortimer and Segal (2005) concluded that brief motivational interventions were cost effective compared with no active treatment among ‘problem drinkers’ whilst MOCE and MET were cost-effective treatments for alcohol dependency, although no common comparators were used in either analysis.

Providing an adequate summary of the health economic evidence presented here is difficult due to the differences across the studies in terms of the interventions and comparators considered, study populations, costs and outcomes considered, and other methodological differences. Overall, the health economic review does not provide evidence of superior cost effectiveness for any particular psychological intervention.

6.21.3. Economic considerations

Of all the psychological interventions included in the systematic effectiveness review and then found suitable for recommendation in the NHS, only a few of these have supporting economic evidence.

A potential solution to this problem would be to undertake economic modelling to determine the most cost-effective psychological intervention. However, certain aspects of the effectiveness evidence made it difficult to do so (that is, there was a lack of common comparators and interventions were usually compared with other active interventions, a ‘no treatment/usual care/placebo’ arm was rarely identified).

Furthermore, the meta-analyses showed that there were small if any differences in effect between active treatments, and only a few of these, for example BCT, showed much evidence of a consistent positive effect, particularly against other therapies.

Therefore, the following costing exercise was undertaken for the possible recommended psychological interventions.

Behavioural therapies

The clinical evidence in the guideline systematic literature review described a variety of interventions that were considered to be behavioural therapies. They were delivered in a variety of sessions and durations either individually or in structured groups under the supervision of a competent practitioner. The clinical evidence was taken in consideration and the GDG agreed that behavioural therapies would typically involve weekly sessions of 1 hour's duration over a 12-week period.

Behavioural therapies can also be delivered by a range of mental health professionals. The GDG highlighted the following professionals: a clinical psychologist, a nurse, or an NHS professional who is trained to deliver this intervention. It was assumed that such workers would be on AfC salary scale 6, which would be likely to be comparable with the salary scales of a community nurse. The unit cost of an AfC Band 6 community nurse is £70 per hour of patient contact and the unit cost of a clinical psychologist is £75 per hour of patient contact in 2007/08 prices (Curtis, 2009). These costs include salary, salary oncosts, overheads and capital overheads plus any qualification costs. Based on these estimates the average cost of a behavioural intervention would be £900 per patient if delivered by a clinical psychologist and £840 per patient if delivered by a mental health professional described above.

A summary of the estimated resource use and costs involved in delivering these psychological interventions is presented in Table 85.

Table 85

Summary of resource use and costs associated with psychological interventions.

6.22.1. Introduction

In the development of the adult treatment sections of this guideline it was accepted that for some people who misuse alcohol (in particular those with harmful use or mild dependence) the reduction in alcohol consumption might be an option. However, given the potential long-term harm experienced by children and young people who are alcohol dependent or harmful drinkers, and the frequent presence of comorbid substance misuse and other psychiatric disorders, it was felt that the appropriate goal for children and young people should be achieving abstinence. However, it was recognised by the GDG that considerable difficulties are faced by some young people in trying to achieve abstinence, particularly if the support they receive from their families, carers and others is limited or non-existent or they experience considerable peer pressure to drink alcohol. Therefore, for some young people the GDG accepted that that an initial reduction in alcohol misuse may be the only achievable short-term objective. Nevertheless, the GDG's view was that, given the considerable problems that young people face, abstinence remained the preferred goal.

A further important difference between the treatment of adults and young people concerns the presence of comorbidities. Although comorbid depressive and anxiety symptoms are common in adults with harmful drinking and alcohol misuse (Weaver et al., 2006), the extent and severity of the comorbidities often found in children is greater (Perepletchikova et al., 2008). Comorbid disorders such as conduct disorder and ADHD significantly complicate the management of alcohol misuse, and concurrent treatment of them is to be considered. This problem is well known (Perepletchikova et al., 2008) and a number of treatments (for example, multisystemic therapy [Henggeler et al., 1999], brief strategic family therapy [Szapocznik et al., 2003] or multidimensional family therapy [Liddle, 1992]) have been developed for conduct disorder explicitly to deal with the complexity of problems faced by children and young people, including drug and alcohol misuse. The latter two interventions have a very explicit focus on substance misuse. At the heart of all these interventions lies the recognition of the considerable complexity of problems presented by young people who misuse alcohol and drugs, and the need often to develop a multisystem, multi-level approach to deliver integrated care.

6.22.2. Aim of review

This section aims to review the evidence for psychological interventions for the treatment of alcohol dependence and harmful alcohol use in children and young people. However, although there are several published reviews on the efficacy of psychological interventions for adults and for the prevention of adolescent substance misuse, there are only a limited number of trials assessing the clinical efficacy of psychological interventions for alcohol misuse alone (without comorbid drug misuse) for children and young people under the age of 18 years. In addition, the patient populations assessed in these trials more often than not have comorbid substance misuse. Therefore, a GDG consensus-based decision was agreed that the literature search would be for alcohol-specific primary studies as well as published systematic reviews to guide the overall strategy of a narrative synthesis of the evidence.

Psychological therapies were considered for inclusion in the review if they were:

• alcohol-focused only
• planned treatment (especially for brief interventions)
• for treatment-seeking participants only (of particular importance for the brief interventions because the scope did not cover opportunistic brief interventions – see Appendix 1)
• manual-based or, in the absence of a formal manual, the intervention should be well-defined and structured
• ethical and safe.

6.22.3. Review questions

The primary review question addressed in this section is:

For children and young people who are alcohol dependent or harmful drinkers, is treatment x when compared with y more clinically- and cost-effective, and does this depend on the presence of comorbidities?

6.22.4. Clinical review protocol

As part of the overall search for effective individual, group and multicomponent psychological and psychosocial interventions for children and young people, the review team conducted a systematic review of published systematic reviews (in part, to take account of the complex comorbidity) of interventions for young people who misuse drugs and alcohol, and also RCTs of interventions for children and young people for alcohol misuse specifically. The literature search identified a number of primary studies investigating the efficacy of psychological interventions for children and young people. However, the participant population in the majority of these studies did not reach inclusion criteria for drinking severity and could not be classified as dependent/harmful. See Table 86 for a summary of the clinical review protocol for the review of psychological interventions for children and young people.

Table 86

Clinical review protocol for the review of psychological interventions for children and young people.

6.22.5. Narrative review

This review of psychological and psychosocial interventions for children and young people should be read in conjunction with the review of brief interventions contained in the NICE public health guidance (NICE, 2010a), and the review of psychological interventions for adults contained within this guideline. A limited number of studies, specifically on alcohol-focused interventions, have been undertaken for children and young people. However, a number of studies have considered the treatment of conduct disorder in the presence of drug or alcohol misuse. In light of this significant comorbidity, in addition to the two guidelines referred to above, the GDG also drew on other recent NICE guidelines, specifically the review of conduct disorders in young people contained within the NICE guideline on antisocial personality disorder (NICE, 2008a) and three other systematic reviews (Perepletchikova et al., 2008; Tripodi et al., 2010; Waldron & Kaminer, 2004). Individual- and group-based therapies and multicomponent interventions used in the treatment of alcohol dependence and harmful alcohol use in children and young people were considered in the review of the evidence.

6.22.6. Evidence summary

The evidence reviewed using these systematic reviews and primary studies suggests that, although there has been recent progress in the development of individual or group psychological treatment of alcohol dependence and harmful alcohol use in children and young people, no individual treatment has a convincing evidence base for harmful alcohol use or dependence. In some respects this is in line with the adult literature and findings of the guideline meta-analyses where a number of structured treatments (including CBT, behavioural therapies and SBNT) had some benefits for harmful and mildly dependent use, but it was not possible to distinguish between them on the basis of the current available evidence. The issue is further complicated by the fact that many of the trials that evaluate the efficacy of these interventions, and that are representative of this population, involved participants with comorbid drug misuse.

6.22.7. Multicomponent psychological interventions

6.22.8. Evidence summary

The evidence for the use of multicomponent interventions demonstrates clear benefits on offending behaviour and promising results for the reduction of alcohol and drug misuse. As was found with the individual- or group-based interventions, much of the research focuses on children and young people with substance-use disorders and who are more likely have comorbid psychiatric disorders. Although not specifically focused on alcohol, this does not significantly detract from their applicability to this guideline because comorbidity with conduct disorder and polydrug use is a common feature among young people with significant alcohol misuse. The research to date does not, however, favour one particular multicomponent intervention over another for the treatment of alcohol misuse.

Children and young people

The evidence base is limited for the treatment of alcohol misuse in children and young people. As a consequence, the GDG was required to extrapolate from a number of datasets and sources that did not directly address the treatment of alcohol misuse in children and young people. This included data on adults with alcohol misuse, as well as children and young people with substance misuse, conduct disorder and antisocial personality disorder. The GDG considered this to be a justified approach because there is an urgent need to provide recommendations for the treatment of the increasing problem of alcohol misuse in children and young people. In extrapolating from these datasets the GDG was cautious, recognising that as new evidence emerges the recommendations in this guideline will need revision.

Despite limited evidence a reasonably clear picture emerged about the effectiveness of interventions to promote abstinence and prevent relapse in children and young people. There was some evidence for individual interventions such as CBT and less so for MET. There was stronger evidence for the use of multicomponent interventions such as multisystemic therapy, functional family therapy, brief strategic family theraphy, and multi-dimensional family therapy, but little evidence to determine whether one of the interventions had any advantage over the others. This evidence also mirrored the evidence for effectiveness in adults. The GDG therefore decided that both types of intervention should be made available with CBT reserved for cases where comorbidity is either not present or of little significance; where comorbidity is present, multicomponent interventions should be offered.

Delivering psychological and psychosocial interventions for children and young people

6.24.1.8.

Multidimensional family therapy should usually consist of 12–15 family-focused structured treatment sessions over 12 weeks. There should be a strong emphasis on care coordination and, if necessary, crisis management. As well as family sessions, individual interventions may be provided for both the child or young person and the parents. The intervention should aim to improve:

• alcohol and drug misuse
• the child or young person's educational and social behaviour
• parental well-being and parenting skills
• relationships with the wider social system.

6.24.1.9.

Brief strategic family therapy should usually consist of fortnightly meetings over 3 months. It should focus on:

• engaging and supporting the family
• using the support of the wider social and educational system
• identifying maladaptive family interactions
• promoting new and more adaptive family interactions.

6.24.1.10.

Functional family therapy should be conducted over 3 months by health or social care staff. It should focus on improving interactions within the family, including:

• engaging and motivating the family in treatment (enhancing perception that change is possible, positive reframing and establishing a positive alliance)
• problem solving and behaviour change through parent training and communication training
• promoting generalisation of change in specific behaviours to broader contexts, both within the family and the community (such as schools).

6.24.1.11.

Multisystemic therapy should be provided over 3–6 months by a dedicated member of staff with a low caseload (typically between three and six cases). It should:

• focus specifically on problem-solving approaches with the family
• use the resources of peer groups, schools and the wider community.

Interventions for harmful drinking and mild alcohol dependence

6.24.1.12.

For harmful drinkers and people with mild alcohol dependence, offer a psychological intervention (such as cognitive behavioural therapies, behavioural therapies or social network and environment-based therapies) focused specifically on alcohol-related cognitions, behaviour, problems and social networks.

6.24.1.13.

For harmful drinkers and people with mild alcohol dependence who have a regular partner who is willing to participate in treatment, offer behavioural couples therapy.

6.24.1.14.

For harmful drinkers and people with mild alcohol dependence who have not responded to psychological interventions alone, or who have specifically requested a pharmacological intervention, consider offering acamprosate³⁹ or oral naltrexone⁴⁰ in combination with an individual psychological intervention (cognitive behavioural therapies, behavioural therapies or social network and environment-based therapies) or behavioural couples therapy (see Chapter 7 for pharmacological interventions).

Delivering psychological interventions

6.24.1.15.

Cognitive behavioural therapies focused on alcohol-related problems should usually consist of one 60-minute session per week for 12 weeks.

6.24.1.16.

Behavioural therapies focused on alcohol-related problems should usually consist of one 60-minute session per week for 12 weeks.

6.24.1.17.

Social network and environment-based therapies focused on alcohol-related problems should usually consist of eight 50-minute sessions over 12 weeks.

6.24.1.18.

Behavioural couples therapy should be focused on alcohol-related problems and their impact on relationships. It should aim for abstinence, or a level of drinking predetermined and agreed by the therapist and the service user to be reasonable and safe. It should usually consist of one 60-minute session per week for 12 weeks.

Why this is important

Psychological interventions are an important therapeutic option for people with alcohol-related problems. However, even with the most effective current treatment (for example, cognitive behavioural therapies and social network and environment-based therapies), the effects are modest at best and the treatments are not effective for everyone. Contingency management has a considerable and compelling evidence base in the treatment of substance misuse (for example, opioid misuse) but there is only a limited, if promising, evidence base for contingency management in the treatment of alcohol-related problems. The results of this research will have important implications for the provision of psychological treatment for alcohol misuse in the NHS.

Introduction

Acupuncture is a form of Chinese medicine that has been practiced for over 3,000 years (Jordan, 2006). It involves inserting fine needles at selected points on the skin to balance the body's energy (chi), with the aim of treating and preventing disease. Acupuncture was introduced specifically for use in the treatment of substance-related disorders approximately 30 years ago (Kao, 1974; Leung, 1977; Sacks, 1975; Wen, 1973). However, research has predominantly been for drug misuse, for example opioid dependence (Jordan, 2006) and cocaine dependence (Gates et al., 2006; Mills et al., 2005), as well as nicotine dependence (White et al., 2006). Research for the use of acupuncture in alcohol misuse is rather more limited and to date there are only two systematic reviews of acupuncture for alcohol dependence (Cho & Whang, 2009; Kunz et al., 2007). Addiction-specific auricular acupuncture involves inserting five small needles on each ear at points regarded to be specific to chemical dependence (known as ‘shen men’, ‘sympathetic’, ‘kidney’, ‘liver’ and ‘lung’) (Smith & Khan, 1988; Wen, 1979).

6.25.1. Clinical review protocol

In the current review, the role of acupuncture in maintaining abstinence and drinking reduction was investigated. Its application to other aspects usually associated with alternative therapies in this topic area (such as craving and withdrawal symptoms) was beyond the scope of this guideline and hence was not considered. Information about the databases searched and the inclusion/exclusion criteria used for this section of the guideline can be found in Chapter 3. The GDG was of the opinion that a search for RCT studies alone may result in an insufficient number of studies to perform a review, therefore a consensus-based decision was made to also search for systematic reviews. See Table 87 for a summary of the clinical review protocol for the review of acupuncture.

Table 87

Clinical review protocol for the review of acupuncture.

6.25.2. Studies considered for review

The review team conducted a systematic search of RCTs and published systematic reviews that assessed the beneficial or detrimental effects of acupuncture in the treatment of alcohol dependence or harmful alcohol use. Following the literature search, 11 primary studies were identified. Of these, four investigated the effects of acupuncture on withdrawal symptoms and two assessed its use for the management of cravings. These six studies were excluded because the outcomes are outside the scope of this guideline. Therefore, five studies (four RCTs and one observational study) were identified for inclusion in the review. However, the review team could not perform an unbiased and comprehensive meta-analysis because there were inconsistent outcome measures across studies. Therefore, the GDG consensus was that a narrative summary of these studies would be conducted. The studies included for review were Bullock and colleagues (1987) (addiction-specific versus non-specific acupuncture); Bullock and colleagues (1989) (addiction-specific versus non-specific acupuncture); Bullock and colleagues (2002) (addiction-specific acupuncture versus symptom-based acupuncture versus non-specific acupuncture versus standard care); Rampes and colleagues (1997) (addiction-specific versus non-specific acupuncture versus no treatment control); and Worner and colleagues (1992) (addiction specific acupuncture versus sham transdermal stimulations versus standard care control). These studies were conducted between 1987 and 2002, and provided data on 752 participants. See Table 88 for characteristics of these studies. All included studies were RCTs except Bullock and colleagues (1989).

Table 88

Summary of study characteristics for acupuncture.

6.25.3. Evidence summary

Bullock and colleagues (1987) investigated acupuncture at addiction-specific points versus non-specific points for reducing craving and maintaining abstinence. The authors reported that the treatment group had significantly fewer drinking episodes than the control group (p = 0.007) after the second (28 days) and third (45 days) phase of treatment, but not after the first phase (5 days).

Bullock and colleagues (1989) also investigated acupuncture at addiction-specific points versus non-specific points for craving reduction, maintaining abstinence and drinking reduction in people with chronic alcohol misuse. The study found that there was no significant difference between the treatment group and control group at 1-month follow-up in the number of drinking episodes (consumption of more than three drinks in one period). However, at both 3- and 6-month follow-up, the treatment group reported significantly fewer drinking episodes than the control group (p <0.001). Furthermore, the treatment group was significantly more effective than control at maintaining abstinence and controlled drinking goals when assessed at 1-month (p <0.01), and at 3- and 6-month follow-up (both p <0.05). This study was not randomised, therefore the results must be viewed with caution.

Worner and colleagues (1992) evaluated acupuncture at addiction-specific points versus needleless transdermal stimulation as well as a standard care group that received no acupuncture. This study found no significant difference between groups in the number of participants who relapsed or needed further withdrawal management at 3-month follow-up.

Rampes and colleagues (1997) assessed addiction-specific electro-acupuncture versus non-specific electro-acupuncture and no treatment (control). The main outcome of interest was craving reduction, which is outside the scope of this guideline. However, the authors also reported no significant difference between groups in amount of alcohol consumed at 2- and 6-month follow-up.

Bullock and colleagues (2002) investigated addiction-specific and non-specific acupuncture as well as symptom-based acupuncture and standard care (based on the Minnesota model). The authors found no significant difference in alcohol consumption at 3-, 6- and 12-month follow-up. Overall, the evidence suggests that acupuncture is not effective in drinking reduction and maintaining abstinence.

The results of these studies are conflicting and show both a benefit of addiction-specific acupuncture as well as no difference between addiction-specific acupuncture and other control conditions. Additionally, the treatments across studies are not comparable because the studies used different body parts for acupuncture treatment, different types of control group, had different length of treatment and follow-up, and varied significantly in sample size. Although the quality of these trials are acceptable in the most part, the number of studies are limited and there is not enough evidence to confirm the benefit of acupuncture in maintaining abstinence or reducing the amount of alcohol consumed. Therefore no clinical recommendations are made but the GDG has developed a recommendation for further research.

6.25.4. Research recommendation

6.26.1. Introduction

There is an increasing recognition that alcohol misuse affects the entire family and the communities in which these families live but what constitutes best practice in the area is not well understood (Copello et al., 2006). What is not in doubt is the considerable suffering and hardship experienced by many families where a family member has a significant alcohol misuse problem (Orford et al., 2005).

In developing this guideline the GDG drew on a previous review of psychological interventions for carers that had been undertaken for the NICE guideline on psychosocial interventions for drug misuse (NCCMH, 2008). This was a pragmatic decision because the previous review had drawn on literature covering both drug misuse and alcohol misuse, and searches conducted for this current guideline had failed to find any substantial new evidence for interventions to support family members and carers. The outcome of the NCCMH (2008) review is summarised below in narrative form.

The NCCMH (2008) guideline identified a number of interventions in the drug and alcohol field that had been developed and tested in formal trials. They are listed below.

6.26.2. Summary of the NCCMH 2008 review

The review identified a total of three RCTs including two trials (Kirby et al., 1999; Meyers et al., 2002) for community reinforcement and family training compared with 12-step self-help groups and one trial (Copello et al., 2009⁴¹) of the five-step intervention in which five-step interventions of various intensities were compared.

Neither Kirby and colleagues (1999), or Meyers and colleagues (2002) found any significant different between community reinforcement and family training and 12-step self-help groups for reported levels of drug or alcohol use⁴². However, Kirby and colleagues (1999) found statistically significant changes from baseline for both groups in relation to carer problems and psychological functioning. In contrast, Meyers and colleagues (2002) found no statistically significant differences (after Bonferroni corrections for multiple testing) in changes from baseline at 12-month follow-up. In the case of the five-step intervention Copello and colleagues (2009) on two primary outcomes related to physical and psychological health and coping, found no statistically significant differences between the full intervention and the guided self-help conditions for both physical and psychological health (WMD −0.23; 95% CI, −4.11 to 3.65) and coping (WMD −0.12; 95% CI, −5.42 to 5.19).

6.26.3. Clinical summary

For both community reinforcement and family training and five-step intervention, there were no statistically significant differences found between these more intensive interventions and self-help (that is, 12-step self-help groups and guided self-help). It appears that self-help interventions are as effective as more intensive psychological interventions in reducing stress and improving psychological functioning for families and carers of people who misuse drugs and alcohol.

6.26.4. From evidence to recommendations

In developing the recommendations for this section of the guideline the GDG also took into account the reviews of family members' experience in Chapter 4 of this guideline which confirmed the view that families typically have considerable unmet needs. This meant that despite the limited evidence the GDG felt that the provision of information and the use of a range of self-help interventions (with relatively low cost) should be offered to families. The GDG also felt that where families could not make use of or have not benefitted from the use of the self-help materials that an offer of a structured intervention as set out in the five-step intervention should be made.

6.26.5. Recommendations

6.26.5.1.

When the needs of families and carers of people who misuse alcohol have been identified:

• offer guided self-help, usually consisting of a single session, with the provision of written materials
• provide information about, and facilitate contact with, support groups (such as self-help groups specifically focused on addressing the needs of families and carers).

6.26.5.2.

If the families and carers of people who misuse alcohol have not benefited, or are not likely to benefit, from guided self-help and/or support groups and continue to have significant problems, consider offering family meetings. These should:

• provide information and education about alcohol misuse
• help to identify sources of stress related to alcohol misuse
• explore and promote effective coping behaviours
• usually consist of at least five weekly sessions.