The Preliminary Evaluation of Supporting Addiction Affected Families Effectively

Rationale:

Well over 100 million family members worldwide are affected by the addictive behaviour (alcohol, drugs, etc.) of a relative, which has a highly stressful impact on these affected family members (AFMs). Although AFMs may not suffer from a diagnosable mental health condition, their experiences of living with a drinker makes them vulnerable to physical and mental ill-health, including mood and substance use disorders, trauma, and stress related conditions. Studies in India on the impact of substance use demonstrate high burden in AFMs, disruptions in family routine, finances, interactions and leisure, and spouses being 'worried about their husband's habits', primarily alcohol consumption. Using evidence-based interventions, AFMs can be helped to reduce their symptoms and improve methods of coping. One such intervention is the 5-Step Method (developed in the UK) and is being used in developed countries, with studies demonstrating significant positive effects. Supporting Addiction Affected Families Effectively (SAFE) is a research project to contextually adapt the 5-Step Method (in India). The project entails a Pilot Randomised Controlled Trial (RCT) evaluating SAFE delivered by lay counsellors in comparison with Enhanced Usual Care (EUC). The study involves preliminary testing of the intervention consistent with the pre-evaluation phase of the Medical Research Council's framework for complex interventions.

Research questions:

The objectives of the study are to evaluate the effectiveness, feasibility and acceptability of the SAFE intervention, for family members affected by a relative's alcohol use, in Goa, India. Corresponding to these objectives, the research questions in the primary research publication are: 1. Primary question: a) Is the SAFE intervention superior to EUC in reducing psychological and physical symptoms in family members affected by a relative's alcohol use, at 3 months' post enrolment? 2. Secondary questions: a) Is the SAFE intervention superior to EUC in reducing coping behaviors in family members affected by a relative's alcohol use, at 3 months' post enrolment? b) Is the SAFE intervention superior to EUC in reducing impact on family members affected by a relative's alcohol use, at 3 months' post enrolment? c) Is the SAFE intervention superior to EUC in improving social support for family members affected by a relative's alcohol use, at 3 months' post enrolment? The research questions in the secondary research publication are: a) What are the treatment experiences of the family members affected by a relative's alcohol use? b) What is the preliminary feasibility of the SAFE intervention, delivered by lay counsellors to family members affected by a relative's alcohol use? c) What is the acceptability of the SAFE intervention, delivered by lay counsellors to family members affected by a relative's alcohol use?

Design:

Parallel arm, single blind, pilot RCT of the SAFE intervention compared with EUC for family members affected by their relative's alcohol use, in Goa; and a nested qualitative study. 50 AFMs each will be recruited in the intervention and control groups.

Study setting and target population:

The study will be conducted in Goa (India; population=1.4 million people). Unlike most of India, Goa has easily available and cheap alcoholic drinks, and a more liberal, 'wet' culture towards drinking and this is reflected in lower abstinence rates. The prevalence of current drinking (past one year) in Goa, amongst men from the community is 39%, amongst primary care male attenders is 59%, and amongst male industrial workers is 69%. The target population includes family members (e.g. parents, spouse/partner, siblings, grandparents) affected by their relative's drinking.

Recruitment:

AFMs will be recruited by 1) referrals from community gatekeepers (e.g. community health workers, self-help group members, village council members), 2) referrals from professionals who come into contact with people with alcohol-related problems or their AFMs (e.g. psychiatrists, priests), and 3) self-referrals resulting from media (e.g. advertisements) and word-of-mouth publicity. Once the AFM is referred or identified, there will be an initial meeting during which the lay counsellors will implement the following procedures: 1) Confirm participant eligibility to enter the trial, 2) Obtain informed consent, 3) Administer baseline assessments, and 4) Assign participants to either SAFE or EUC, based on the randomisation list.

Baseline measures:

Baseline measures include: 1) Socio-demographic questionnaire, 2) Symptom Rating Test (SRT): assessing the extent of mild-to-moderate physical and psychological ill heath, 3) Coping Questionnaire (CQ): measuring ways of coping in the AFM, 4) Family Member Impact Questionnaire (FMI): measuring extent and type of impact on the AFM, and 5) Alcohol, Drugs and the Family Social Support Scale (ADF-SSS): assessing the perceived functional social support needs of AFMs. The SRT, CQ, FMI, ADF-SSS have been selected as: they have been used in other studies exploring the 5-Step Method, have developed from decades of work with AFMs, and measure the key underpinning theoretical constructs in the Stress-Strain-Coping-Support model on which the 5-Step Method is based. All these measures have been validated previously and will have undergone adaptation for use in the local context during the ongoing formative research.

Randomisation:

Consenting eligible participants will be randomized to receive the intervention or EUC. An independent data manager will generate a randomisation list to avoid any bias in treatment allocation. Research assistants will use Sequentially Numbered Opaque Sealed Envelopes to randomize the participants individually, to maximize allocation concealment.

Intervention and Enhanced Use Care:

The intervention to be tested will be a contextually adapted version of the 5-Step Method, which is based on the principles of the Stress-Strain-Coping-Support model. The 5 steps include: 1) Exploring stresses and strains, 2) Providing relevant information, 3) Exploring and discussing coping behaviors, 4) Exploring and enhancing social support, and 5) Ending sessions and exploring additional needs, and further sources of help. The intervention will be delivered in settings based on convenience of the participant (home, health centre, etc.). In the study setting, usual care for AFM is no care at all as detection rates are extremely low. EUC for the control group includes an information leaflet on the nature and impact of alcohol use, and existing treatment options for AFMs.

Interventionists:

SAFE will be delivered by lay counsellors, i.e. community members with no mental health qualifications. Eligible lay counsellors will undergo rigorous training and supervision in the delivery of the intervention and only those who achieve pre-determined competency standards will be selected to deliver the intervention.

Outcomes and process indicators:

The effectiveness of the intervention will be assessed at 3 months, through the primary outcome measures (SRT), and secondary outcome measures (CQ, ADF-SSS and FMI), administered by research assistants. Process indicators will be collated including number of patients screened and completing the intervention; number, frequency and duration of intervention sessions; and number of lay counsellors trained and achieving competency.

Nested qualitative study:

The qualitative study is to explore AFMs' experiences of SAFE and its perceived impact, and lay counsellors' experiences in delivering SAFE. The primary methods of qualitative data collection will be in-depth interviews and focus group discussions. The participants for the qualitative study will be recruited after the completion of the end-point assessments of the study, utilising a purposive and maximum variation sampling strategy.

Strategies to minimise bias:

The investigators and research staff will be blind to the treatment allocation. The following procedures will be implemented to ensure blinding: 1) The 3-month outcome assessments will be administered by research assistants who will have had no previous engagement in the study, and will be 'blind' to the treatment allocation, to avoid bias in the assessment of the outcomes. 2) To minimise unmasking: a) the primary outcome assessment will be administered prior to all other assessments; b) during the training of the research assistants, it will be emphasised that there is genuine scientific equipoise about the intervention versus EUC; c) participants will be requested not to disclose their allocation status to the research assistants during the assessment. 3) The baseline data will be anonymised during the analysis stage. During the study, the blind may be broken in circumstances where the participant experiences any harm/risk of harm, and knowledge of the treatment allocation status is crucial for further management.

Analyses:

Baseline comparability of randomized groups: Baseline characteristics of enrolled participants will be compared between treatment arms and overall, summarized using mean and standard deviation, median and interquartile range or numbers and proportions as appropriate.

Outcome analyses: STATA will be used for data description and the main inferential analysis. The primary analyses will be intention-to-treat at the 3-month end-point. Any missing outcome data will be imputed by multiple imputation, using baseline characteristics on STATA. Secondary analyses at the 3-month end point be complete case analyses adjusted for baseline values associated with drop out from outcome evaluation.

Process evaluation: Process evaluation will be used to assess quality of implementation in the trial, and identify contextual factors associated with variation in outcomes. Process indicators will be summarized in a CONSORT flow chart which will include recruitment and consent rates. Initial analyses will compare baseline characteristics of individuals who consented and did not consent, and participants who completed outcome assessments and did not complete assessments. The routine monitoring data will be summarised in a descriptive manner.

Qualitative analysis: The qualitative data from interviews will be coded independently by two researchers, and themes will be identified and further analysed using the thematic analysis framework.

Overall, the data arising from the outcome and process evaluations will be triangulated for a comprehensive understanding of the study data.

Host institute:

Sangath (Goa) is the coordinating centre for the study (http://www.sangath.com/index.php). Sangath is a community-based Non-Governmental Organisation, working across various states in India, on research in mental health problems across the lifespan. Sangath has achieved global recognition in mental health research, given its focus on innovative solutions in mental health care in low resource settings; and has a track record of conducting successful psychosocial intervention-based RCTs.

Ethics:

The study is planned, conducted and reported in accordance with the Helsinki Declaration and Good Clinical Practice guidelines. Ethical approval has been sought from the Sangath Institutional Review Board (IRB) and the Indian Council of Medical Research (ICMR).