Aims of the study
To explore through qualitative interviews, the acceptability and impact of the initial approach, assessment and attempted recruitment on participants and non-participants in the trial.
To explore through qualitative interviews, the acceptability of the healthy living intervention and the alcohol-focused intervention in a population of non-help seeking problem drinkers and alcohol-dependent patients presenting in the hospital setting.
To compare the clinical and cost-effectiveness of a healthy living intervention and an alcohol-focused intervention in their impact on drinking problems.
To compare the effects of a healthy living and alcohol-focused intervention on: quantity and frequency of drinking, dependence, social satisfaction, psychological health and health-related quality of life.
To evaluate the level of treatment retention amongst participants randomized to the two interventions, based on numbers attending first and subsequent sessions.
To explore the acceptability and feasibility of postal assessment instruments at 6- and 12-month follow-ups.
To establish patient preferences between study interventions prior to treatment.
All patients admitted to the selected wards are considered potential participants. The hospital in-reach team identifies patients who have an alcohol-related diagnosis or a reason for admission that suggests a possible alcohol-related diagnosis, which is thought to be responsive to addiction interventions.
Eligibility is assessed using a screening form, which includes the AUDIT plus a checklist for the other eligibility (inclusion/exclusion) criteria. Patients with an AUDIT score ≥16 who meet the remaining eligibility criteria are provided with a patient information sheet (PIS). All patients with an AUDIT score ≥16, but who do not meet the remaining eligibility criteria continue with their care as current practice dictates.
Patients scoring 15 or less on the AUDIT who have been admitted with an alcohol-related problem are given a leaflet containing information on units, stopping or reducing alcohol intake, the support available at a local specialist service, the address of an interactive website designed to guide people towards reducing drinking and are offered treatment as usual. The leaflet is based on Department of Health advice and evidence from previous trials including the Screening and Intervention Programme for Sensible Drinking trial
Completed screening forms, including anonymous ones for those not eligible or not willing to enter the trial, are returned to the coordinating center. This information is used to collate eligibility and non-eligibility or non-participation information, as well as providing AUDIT scores, reason for admission, age and gender for prevalence data and to inform future studies.
Potential participants are provided with a verbal explanation of the study and a copy of the PIS. The PIS outlines the purpose of the study, the proposed interventions, study processes and time commitment required for both treatment and follow-up assessments. For those agreeing to participate, written consent is obtained and a baseline questionnaire is completed. An additional PIS gives information and asks for consent to participate in the qualitative element of the study.
Six therapists are divided into three pairs, broadly matched on experience and qualifications. Within each pair, one therapist is trained to deliver the alcohol-focused intervention and the other trained to deliver the healthy living treatment. Each participant is randomized such that they have a 50:50 chance of getting the therapist who screened them or the matched colleague. This approach is based on previously obtained opinions from service users stating that continuity of care is important to them. Should one of the therapists in the pair not be available due to illness or similar reasons, the participant is allocated to a therapist within a different pair at the point of randomization. Participants are randomized by a secure remote computer service run by the fully registered York Trials Unit, to either an alcohol-focused intervention (control treatment) or a healthy living intervention.
Both interventions are delivered in four sessions using motivational dialogue based on the guiding principles of motivational interviewing
. This number of sessions has been determined using evidence from the UK Alcohol Treatment Trial (UKATT), which suggest optimal attendance is achieved over four sessions
Each intervention consists of four sessions of 30 to 45 minutes each, delivered one to two weeks apart, over a maximum period of 8 weeks. Sessions for either arm can be delivered at the specialist clinic, the participant’s home or at a mutually agreed suitable alternative location as is current practice for the Hospital In-Reach Team (HIRT).
Both interventions involve prompting the formation of intention through encouraging participants to set a general goal or make a behavioral resolution, and encouraging them to set specific goals, self-monitor behavior using a diary. Feedback from the participant on their performance allows them to review the behavioral goals together with the therapist. Motivational dialogue is used to elicit behavior change by helping clients explore and resolve ambivalence and facilitate treatment goal-setting
The alcohol-focused intervention (iSBNT) is a social-network-based, cognitive behavioral intervention using motivational dialogue
 and drawing on cognitive behavioral therapy (CBT), which has been shown to be an effective approach for treating patients with alcohol problems
[25–27]. It has a strong evidence base in UK clinical trials and is increasingly commonly taught and practiced in the UK addiction field. It contains exclusively drinking behavior change topics and involves strengthening coping skills and confidence, with emphasis on structure and skill rehearsal. Behavioral rehearsal is used incrementally to develop the service user’s and network members’ skill levels and confidence to apply new skills in their own environment. Homework is used to implement new practice and facilitate achievement of agreed treatment goals.
The participant is encouraged to recruit a supportive network of concerned others, who will help the participant to implement behavior change and develop new lifestyle habits inconsistent with drinking. The network is encouraged to make plans to cope with relapse risks and actual relapse, providing support for each other where necessary.
The healthy living intervention, which is also a behavioral approach, combines behavior change in up to three chosen domains that contribute to overall health. Seven options from which the participant chooses three are diet, exercise, drinking, smoking, medication concordance, drug use and personal care. Choice is a key variable in the study as it is the basis of exploring acceptability
. The participant is encouraged to enlist the help of someone who is concerned about them and broadly shares their behavioral goals. At the first session, the target domains are agreed. These are not set in stone but are likely to be the ones to be pursued. The participant is then asked in which order they wish to pursue the domains and a goal is set for the first one. Behavioral plans are made for achieving the goal and homework is set to put these plans into practice. At the second session, progress towards the previously set goal is explored and reviewed and the next domain is addressed in the same way. This continues in the third and fourth sessions. Where the behavioral goals for a domain are not achieved, the therapist and participant decide whether to repeat or to move on to the next domain. The participant is encouraged to identify and bring along a buddy to support the intervention and this can be the same person or separate individuals for each domain. Drinking does not have to be one of the chosen domains.
Concurrent treatment can occur wherever there is a need for medical treatment of a physical condition, or surgical or psychiatric treatment that does not preclude delivery of the study interventions. If after their participation in the research is complete, a participant wishes to continue with specialist addiction treatment, they can request this and their referral will be treated in the normal way.
Six addiction practitioners working in the hospital in-reach team received dedicated training covering the study’s procedures and the practices and techniques of each intervention, followed by supervised practice of study treatments until deemed competent. The training package is delivered separately to the two treatment groups by an experienced trainer/addiction psychologist. Once deemed competent, practitioners continue to deliver the manually guided study interventions under regular supervision based on recorded practice to avoid therapist drift. Additional training is given on the study protocol process, documentation completion and obtaining informed consent.
Due to the nature of the interventions in this study, blinding of participants and therapists to allocated treatment is not possible. However, baseline data were collected prior to randomization and follow-up data will be collected by postal questionnaires: those involved in the analysis of the data are blind to treatment allocation.
Although assessing the components and processes of the design is an important feature of this pilot, the main purpose is to evaluate the acceptability of the interventions to the participants. Participant interviews are being undertaken to explore the acceptability and impact of approach, assessment, recruitment and intervention on trial participants and non-participants (those not willing to take part in the trial, but do consent to an interview). A purposive sample of 20 to 30 eligible participants are being interviewed face-to-face while still in hospital – both those who have expressed an interest in taking part in the trial and non-participants, exploring their thoughts and feelings concerning their identification as a problem drinker in the hospital context and how this status compares to their own perceptions of their drinking behavior. This also allows us to explore the impact of hospitalization and related health concerns on motivation to engage in treatment. The interviews with the non-participants also probe reasons for not participating in the study.
Semi-structured interviews (mixture of face-to-face and telephone) are also being undertaken with a sample of 10 to 15 individuals from each arm of the trial who have participated in part or all of their treatment sessions. These interviews examine a broad range of issues including their responses to the baseline assessment, how the social engagement functions (integral to both interventions) have worked, the perceived obstacles to change and how formal elements of the interventions have interacted with other aspects of their lives, including significant life events and changes of circumstances. The data from these interviews will add to the quantitative measures in addressing the acceptability of these interventions and the possible mechanism of their impacts. Importantly, they ensure that patient perspectives are included in the evaluation of these interventions.
In addition, the therapists delivering the interventions will be interviewed once recruitment is well under way to ask them about their experience of identifying people with alcohol issues and how patients responded to their approach. The therapists will be provided with an information sheet and written consent will be obtained.
All interviews are being conducted by an experienced mixed-methods researcher using the appropriate topic guide/ schedule for the sample being interviewed (that is, in hospital, post treatment or therapist) to promote consistency. Where permitted, interviews are audio-recorded. In addition, a second researcher will independently perform a thematic analysis on a sub-sample of transcripts, with discussions taking place to develop an agreed coding framework.
Patient outcome measures
This study also provides an opportunity to compare the effectiveness of the two interventions. Therefore, the AUDIT score at 6 months post randomization has been designated as the main patient outcome and is also measured at 12 months post randomization, whilst the proportion of participants who have not consumed any alcohol in the previous 6 months is collected via a yes/no question at 6 and 12 months post randomization.
A number of secondary patient outcomes are also measured. The Leeds Dependence Questionnaire (LDQ)
, a ten-item, self-administered measure designed to assess severity of dependence is helpful in determining a patient’s treatment goals
. Social satisfaction is measured using the Social Satisfaction Questionnaire (SSQ), a validated eight-item scale found to be a suitable outcome measure in people with substance use disorders covering social domains that are likely to be important to, and engage, service users in treatment
Subjective well-being, psychological problems and functioning are measured using the Clinical Outcomes in Routine Evaluation (CORE-10)
. The health-related quality of life (HRQoL) is assessed using the European Quality of Life – 5 Dimensions (EQ-5D). The EQ-5D is a standardized measure of health status developed by the EuroQol Group in order to provide a simple, generic measure of health for clinical and economic appraisal, where health is characterized on five dimensions (mobility, self-care, ability to undertake usual activities, pain and anxiety/depression)
. These four measures are taken at baseline, and 6 and 12 months post randomization.
At the end of the second and third treatment sessions, both participant and therapist are asked to complete the 12-item (short version) working alliance inventory (WAI)
 in order to assess the participant–therapist relationship.
The costs of delivering the two interventions are calculated on the basis of resources used. Participant use of health services, other alcohol services outside the study, public services and criminal justice services is assessed through service use-questions at baseline, and 6 and 12 months post randomization. We will use utility values based on societal values using the York tariff
Recruitment rates are presented in three tiers: number screened, number eligible and number participating, with the last of these used to infer initial acceptability. Retention in treatment is evaluated by the number of sessions attended and used as a measure of treatment acceptability. Quantitative assessment of the acceptability of the postal follow-up questionnaires is assessed by return rates and completion rates. Qualitative assessment of any issues regarding questionnaire completion is explored during the interviews. Patients’ reported preferences for either or neither of the study interventions are collected in the baseline questionnaire, prior to randomization.
All follow-up data are collected by postal questionnaire at 6 and 12 months post randomization with postal reminders sent 2 and 4 weeks after initial correspondence. In the event of a participant failing to complete their questionnaires, attempts to obtain at least the primary outcome data are made via telephone.
Quality assurance of treatment delivery
All treatment sessions are video-recorded (the participant is not visible) for the purpose of supervision in order to prevent practitioner drift. These are independently evaluated to rate the delivery of the essential and unique components of each treatment using an instrument adapted from the UKATT process rating scale
The main aims of this pilot study are to evaluate the acceptability of the interventions to the participants, and the elements and processes of the design. Using the opportunity available to compare the effectiveness of the two interventions and advising whether one or both could be taken forward into a large-scale study, the planned analyses cover all aspects.
Interviews are digitally recorded (where consent is given), fully transcribed and imported into Atlas.ti
 for the management and analysis of data. Thematic analysis will be used to identify themes by their frequency, intensity and extensiveness, a technique described by Braun and Clarke
Primary statistical analysis
The primary patient outcome measure for the study is the AUDIT score measured at 6 months. Analysis will be on an intention-to-treat basis using two-sided, 5% significance and will compare outcomes for the alcohol-focused intervention with outcomes for the healthy living intervention at 6 months.
The AUDIT scores at 6 months will be analyzed using a linear regression model. The baseline AUDIT score will be included as a covariate and the model will also include the treatment group. Model checking will be performed to ensure that the model is a good fit for the data. If necessary, transformations will be used to improve model fit.
Secondary statistical analyses
Secondary outcomes include the AUDIT score at month 12. They also include LDQ, SSQ, CORE-10 and EQ-5D all measured at 6 and 12 months. These will be analyzed separately for each time point using linear regression models, adjusted for the baseline score of the dependent variable and include a variable for treatment group. All analyses will compare the alcohol-focused intervention with the healthy living intervention.
Abstinence, a binary outcome (yes/no) collected at month 6 and month 12, will be analyzed using logistic regression models for each time point. Included in the models will be a variable for treatment group and baseline AUDIT score.
Preference for treatment will be collected prior to randomization and will be summarized overall and by treatment group. The primary analysis will be repeated but with the addition of the variable for treatment preference and a variable for the interaction between preference and treatment group to examine the effect preference may have on outcomes.
The relationship between client and therapist as measured by the WAI will be explored using descriptive statistics and correlation where appropriate. We will undertake an exploratory analysis to investigate the effect of adherence to treatment using a complier average causal effect analysis
[39, 40]. The numbers of adherent participants and further details of adherence will also be summarized.
The economic component of the study comprises a cost-effectiveness and cost-utility analysis, and aims to identify, quantify and value health-care resources utilized by participants in each trial arm.
Resources utilized in the identification and delivery of the alcohol-focused and healthy living interventions will be recorded. This will allow the calculation of costs related to the implementation of these interventions. Local costs will be used to calculate the costs of the interventions. In addition, specific training costs for staff are calculated in terms of staff time, costs for premises and the cost of training materials.
Participants’ use of health care will be identified retrospectively using service use questions and applying a common set of national unit cost estimates
[41, 42]. Participant costs in the 6-month period before the intervention can then be compared to costs in the 6-month period after receiving the intervention and used to explore any changes in costs imposed by participants in each group.
The economic analysis will calculate the incremental cost-effectiveness of the alcohol-focused intervention compared to the healthy living intervention, using measures of clinical outcome and quality of life (EQ-5D
) responses at baseline and at 6 and 12 month follow-up. The use of EQ-5D enables the estimation of quality-adjusted life years (QALYs). Data will be bootstrapped to account for the expected skewness evident in economic cost data. The analysis will construct cost-effectiveness acceptability curves to illustrate the probability that the alcohol-focused intervention is more cost-effective than healthy living, based on different monetary values being attached to the QALYs. The use of QALYs follows the recommendations of the National Institute for Health and Clinical Excellence
 and enables the value for money afforded by treatment to be compared to a range of other health-care interventions.
Treatment fidelity analyses
Visual recording of all treatment sessions is used to rate the quantity and quality of the treatment components delivered. This rating is based upon an instrument adapted from the UKATT process rating scale
 and performed by a researcher trained in process rating. Correlational analysis of the data derived from rating is performed to detect protocol adherence and discriminability between the treatments.