Alcohol Control in Smoking Cessation

Co-use of Alcohol and Tobacco Co-use of tobacco and alcohol is common globally. Positive association of smoking and drinking at individual level has been found in both community settings and psychiatric epidemiological studies. In the U.S (2001-2002), about 21.7% of current drinkers were also current smokers, and about 34.8% of current smokers were current drinkers. In Hong Kong, according to the population-based study conducted by Hong Kong Council on Smoking and Health in 2018, about 22.2% and 9.3% of current smokers drank at least once a week and drank alcohol daily, respectively. From the client data (2014-16) of Tung Wah Group of Hospitals Integrated Centre on Smoking Cessation (TWGH-ICSC), the largest cessation service provider in Hong Kong, the prevalence of lower-risk drinking, increasing risk drinking, harmful drinking, and probable alcohol dependence, screened by the Alcohol Use Disorders Identification Test (AUDIT). in their clients is about 46.7%, 10.9%, 2.7% and 2.4%, respectively.

Co-use of tobacco and alcohol is harmful and incur substantial economic loss. Both tobacco and alcohol were classified by World Health Organization (WHO) as a Group 1 carcinogens. Concurrent use of tobacco and alcohol leads to 25.3% mortality due to coronary heart disease, stroke, respiratory and smoking-related cancer, compared with 20.5% among people who only smoked and 12.8% among who only had 15+ units of alcohol each week. Co-use of alcohol and tobacco also poses a severe economic burden on family and community. According to the Lancet's 2017 study, tobacco use costs global economies more than US$1 trillion annually through loss of productivity and health-care expenditure, while the weighted average cost of alcohol consumption ranged from 0.45 and 5.44% of Gross domestic Product (GDP)purchasing power parity (PPP) globally.

How Alcohol Use Hinders Smoking Cessation Alcohol consumption hinders smoking cessation, which can be explained by the theory in subjective reward effects, cross-substance cue reactivity, and other psychosocial factors. Based on the subjective reward effects, alcohol consumption increases nicotine's subjective rewarding effects, and hence increases satisfaction, liking, and calming feelings due to nicotine. Therefore, attempts to quit smoking are less successful among those with alcohol use disorder. The cross-substance cue reactivity suggested that there are cognitive or semantic associations between alcohol and cigarette use. That is, one substance will cue the use of the other substance. When alcohol and tobacco are used together for a number of times, the use of either substance will trigger a craving of another substance. This theory was commonly used to explain why alcohol is always an obstacle in smoking cessation and a cue for smoking relapse. These mechanisms have been supported by intervention trials and observational studies, showing that smoking rates among drinkers of different levels were significantly higher compared with non-drinkers after receiving smoking cessation treatment. If health care professionals (HCPs) do not handle smokers' drinking problems, they may miss the opportunity to help smokers to tackle the drinking problems at the same time.

Local Findings From the 2014-16 record of TWGH-ICSC, the self-reported quit rate at 26-week for clients with baseline AUDIT score 1-8, 9-15, and ≥16 after receiving the standard smoking cessation treatment was 43.4% (95% Confidence Interval (CI) 41.8%-45.1%), 37.1% (95%CI 33.8%-40.5%) and 26.7% (95%CI 22.4%-31.4%), respectively. The 52-week quit rate in these three drinking groups showed similar contrast (AUDIT score 1-8: 38.7%, 95%CI 37.0%-40.3%; 9-15: 31.9%, 95%CI 28.7%-35.2%; ≥16: 21.6%, 95%CI 17.7%-26.1%). Therefore, higher alcohol dependence is associated with lower abstinence in quit attempters who receive smoking cessation treatment. Two participants who self-reported tobacco abstinence for over a month from our ongoing qualitative study (funded by Health and Medical Research Fund (HMRF), ref no. 15163001) of smoking cessation stated that they were only smoking when they were drinking alcohol. They expressed that they smoked whenever they were drinking with friends and did not perceive occasional smoking during drinking episode as smoking relapse. All these results suggested that alcohol use reduces the effectiveness of smoking cessation treatment.

Interventions for Smoking and Drinking Co-use The above literature review suggested that a smoking cessation treatment that simultaneously addresses both alcohol and smoking problems is needed in clinical practice. The U.S. clinical practice guideline of treating tobacco use and dependence mentioned that helping smokers to recognize the dangerous situations of alcohol use is essential, and HCPs should instruct smokers to reduce/abstain from alcohol while quitting smoking. Previous study in alcohol treatment showed that alcohol users generally accept a concurrent treatment of both smoking cessation and alcohol.Combination of pharmacotherapy (i.e. varenicline and naltrexone) has been recommended to effectively and simultaneously treat heavy alcohol use and nicotine dependence.

Evidence-based brief intervention for reducing alcohol use has been established and is recommended by WHO. Behavioral interventions included personalized feedback, decisional balance, quit day setting and tailored skills-building. However, the effectiveness of these behavioral intervention for increasing tobacco abstinence and reducing alcohol use across the intervention trials is inconsistent. The investigators found 3 randomized controlled trials (RCTs) showed a promising effect of the intervention in increasing tobacco abstinence, but other 2 RCTs did not or even reduce tobacco abstinence rate. To late, there has been no systematic review on the effect of incorporating alcohol intervention into smoking cessation treatment. Therefore, the evidence is not yet conclusive.

Local Services for Alcohol Intervention Hospital Authority and non-governmental organizations are providing alcohol treatments, including Hospital Authority's Psychiatry Specialist Out-patient Clinics (SOPCs), Tung Wah Group of Hospitals Integrated Centre for Addiction Prevention and Treatment (ICAPT), TWGH's Alcohol Abuse Prevention and Treatment Service--"Stay Sober, Stay Free," and Alcoholics Anonymous Hong Kong. There has been no publication and data on the utilization of these services.

The Department of Health promoted screening and subsequently delivering a tailored health advice for alcohol users with AUDIT. It published a guide for HCPs to assess drinking and deliver an AUDIT-based brief alcohol intervention in primary health setting in 5-10 minutes. The first three questions of the AUDIT are grouped as AUDIT-C version, which attempts to quickly screen drinkers with high alcohol consumption level. If patients score 3 or above in AUDIT-C, HCPs will proceed to assess alcohol dependence with the rest of the AUDIT. The 3-rd question of the AUDIT-C ("How often do participants have at least 5 cans of beer/5 glasses of table wine/5 peg of spirits on one occasion?) is an indicator of binge-drinking.

According to the local guide on using AUDIT-based intervention, patients with lower-risk drinking (score 1-7in the AUDIT) should receive advice on health risks of drinking and they should drink less or abstain from alcohol. Patients with an increasing risk of drinking (score 8-15 in the AUDIT) should be warned on the increasing risk and impairment due to drinking, and they should cut down drinking. Patients with harmful drinking (score 16-19 in the AUDIT) usually have a pattern of alcohol use that is causing damage to health, hence they should receive a more severe warning and be encouraged to seek professional support. Patients who are defined as probable dependence (score 20 or above in the AUDIT) showed a cluster of behavioral, cognitive, and physiological phenomena, such as strong desire to consume alcohol, impaired self-control, persistent drinking, and physical withdrawal reaction). The HCPs should let them know that drinking is already causing alcohol dependence, and advise the patients to stop or reduce drinking under the help of professionals. It's best to refer patients in this category to a specialist for more in-depth evaluation and treatment. In addition to the above framework, pregnant women, persons on medication and persons with mental illnesses should be asked to abstain completely from alcohol.

Although this brief guide includes the instruction to educate drinkers the risks of excessive alcohol use and refer other services, there are several implementation difficulties. The guide was solely developed for alcohol reduction/abstinence, but it does not involve the intervention for alcohol and tobacco co-use. This guide does not address alcohol use as a barrier of quitting smoking. It does not instruct HCPs on aspects of beliefs, attitude and more personalized advice for smokers. The effectiveness and feasibility in adapting this AUDIT-based in clinical setting has not been evaluated. To address these limitations, an AUDIT-based brief alcohol intervention that can be specifically incorporated in smoking cessation treatment is needed. An alcohol intervention protocol customized for co-users, and training for HCPs about the knowledge, attitude, and practice of delivering alcohol and tobacco intervention are needed.

Proposed Alcohol Intervention In the proposed intervention, AUDIT will be tentatively adopted to screen potential participants. A need assessment will be done to guide the adaptation and revision of the final protocol. To increase the feasibility of applying the alcohol intervention, the need assessment is required to understand (1) current routine in smoking cessation treatment; (2) HCPs' capacity (knowledge, attitudes and practice) and acceptability of the intervention; (3) current co-users' expectations and acceptability of the intervention.

The aforementioned framework in delivering personalized advice to drinkers with lower-risk, drinking at increasing risk, harmful drinking and drinking dependence will be used. Co-users of tobacco and alcohol should be advised the importance of quitting smoking and reducing/abstaining alcohol. The investigators propose a short and brief interventions for co-users who are receiving smoking cessation treatment, because the investigators believe their primary goal is smoking cessation. The co-users may not have enough time or readiness to discuss the issue of drinking problems intensively. A brief advice on drinking may be more acceptable than an intensive intervention for these smokers.

The alcohol intervention protocol will be guided by the theory of self-determination and general health promotion action. The self-determination theory values the importance of internalized autonomy rather than external factors (e.g., requests from family members, friends, or healthcare professionals) during behavioral regulation. Previous RCT had revealed that participants who had the option to decide on their treatment showed a higher level of compliance with instructions. The positive association between autonomy and competence has also been shown by our previous RCT on smoking cessation. Compared with people with less autonomy in treatment, people with greater independence reported greater competence in achieving behavioral change.

The investigators' general health promotion action (GHPA) model stated that people with general intention to promote health are more likely to adopt specific health promotion actions. Specifically, smokers who have general intention to promote health are more likely to stop smoking. If co-users have the intention to improve health, they may pursue both alcohol reduction/abstinence and smoking cessation. One intervention RCT found that smokers who initiated a quit attempt could reduce alcohol consumption by 2.3 units per week after the target quit day, and heavier drinking smokers presented a more significant decrease (6.1 units) in alcohol intake. Based on this model, our proposed study shall encourage the HCPs to increase participants' awareness and intention for health and a healthy lifestyle.

Aims and Hypotheses to be Tested:

This study aims to develop a theory-driven, brief alcohol intervention protocol for HCPs to deliver the intervention for co-users of tobacco and alcohol during smoking cessation treatment. The primary hypothesis is that smoking cessation treatment including the alcohol intervention will increase tobacco abstinence and reduce alcohol drinking. Our second hypothesis is that the brief intervention to deal with alcohol use will be feasible during smoking cessation treatment and be widely accepted by smokers.

Setting The recruitment sites will be the smoking cessation clinics or services under TWGH-ICSC in Hong Kong, which offer behavioral and/or pharmacological interventions. The TWGH-ICSC operates 5 service centers, serving around 3000 smokers each year. The investigators have successful experience of collaborating with these service providers in recruiting smokers for several projects. For example, the investigators recruited around 1,000 smokers from these service providers in an HMRF-funded project in 2018-2019 (Project No. 15163001).

Procedures The HCPs who deliver the smoking cessation counselling will screen the eligibility of the clients at the first or later follow-up of their smoking cessation treatment, and invite eligible participants to join the RCT. When a counselor identifies a potential client, he/she will briefly introduce the alcohol control program. Participants who agree to join the RCT will be required to sign the consent form. HCPs will then randomize the client to either the intervention group or control group afterwards. In the intervention group, participants will receive the standard smoke cessation service and personalized advice on alcohol control. The details and key components of the intervention will be addressed in the intervention section below. In the control group, HCPs will deliver the standard smoke cessation service that same as the usual practice.

Randomization, Blinding The Qualtrics (an online platform for collecting data) randomizer function will be used for the randomization (with 1:1 allocation ratio) in this study. After participants consent to the study, HCPs will access Qualtrics for randomization. The Qualtrics will display the group allocation (either intervention group or control group) after HCPs enter the participants' names. HCPs will inform participants regarding their group allocation immediately. All HCPs and the participants will not be blinded to the given interventions. Outcome assessors at the telephone follow-up are blinded to the group allocation.