Can Brief Email Guidance Enhance the Effects of an Internet Intervention for People with Problematic Alcohol Use? A Randomized Controlled Trial

Abstract

Background

Some research suggests that internet interventions aimed at people with problematic alcohol use are more effective when provided with guidance from a therapist or coach.

Purpose/Objectives: This trial intended to compare the effects of a previously evaluated internet intervention for people with problematic alcohol use when delivered with or without brief email guidance. Methods: Using online advertising, 238 participants, 18 years or older, were recruited and randomized to receive access to the Internet intervention Alcohol Help Center with or without brief email guidance from a health educator. The guidance consisted of at least four structured, slightly individualized emails delivered during the first two weeks after randomization. Participants were followed up at 3 and 6 months. Results: Number of log-ins did not differ significantly between groups throughout the follow-up period. The follow-up rate at 6 months was 47.0%. Generalized estimating equations run on the primary (standard drinks in preceding week/heavy drinking days in preceding week) and secondary outcome variables (AUDIT, AUDIT-C, quality of life) revealed no significant differences between the interventions on any of the outcomes. Conclusions/Importance: The study does not provide support for any added benefits of providing brief guidance via email in an internet intervention for problem drinkers.

Keywords:

• Alcohol
• internet interventions
• email
• randomized controlled trial
• guidance
• cognitive behavior therapy

Introduction

Internet interventions for problematic alcohol use are effective in reducing alcohol consumption in a range of populations, generally rendering small effect sizes (Sundström et al., 2016). These interventions come in two formats: either as a single-module electronic screening and brief intervention (eSBI) providing feedback on the user’s alcohol consumption along with tips on how to reduce drinking (Donoghue et al., 2014), or as cognitive behavior therapy (CBT) modules focusing on goal setting, risk situations, craving and relapse prevention (Hadjistavropoulos et al., 2019). As these interventions exponentially increase access to anonymous help for a population well known to avoid help-seeking due to stigma and embarrassment (Cunningham et al., 1993), identifying ways to maximize their impact is paramount.

Research has shown that providing access to human guidance when using an internet intervention tends to increase both user engagement and effects. Primarily, this evidence has stemmed from the adjoining fields of anxiety and depression, where meta-analyses have shown guided interventions to be more effective than unguided ones (Johansson & Andersson, 2012; Richards & Richardson, 2012) and that these effects stand regardless of the training of the person providing the guidance (Baumeister et al., 2014). Although individual studies have indicated that human guidance may be beneficial also in the context of alcohol interventions (Blankers et al., 2011; Sundström et al., 2016), the first meta-analysis on internet alcohol interventions investigating this issue showed no differences (Riper et al., 2014). However, a recent individual patient data meta-analysis including 19 randomized trials, of which 6 included guided groups, demonstrated that human guidance also had an effect in this population, with individuals in guided interventions consuming a mean of 6.8 standard drinking units less than those in the unguided conditions at follow-up (Riper et al., 2018).

There is no clear, agreed upon definition of human guidance in the literature; format and amount of guidance provided in studies vary. Most commonly, guidance is provided as text messages delivered digitally via the platform through which the intervention is delivered: Blankers et al. (2011) offered seven 40-minute text-based chat therapy sessions over 7 weeks, Sundström et al. (2016) offered either seven 40-minute text-based chat therapy sessions or secure messaging over 10 weeks, Postel et al. (2010) offered 1–2 therapist interactions per week through secure messaging over 3 months, and Boß et al. (2018) offered secure messaging over 5 weeks. In one study, guidance was delivered completely through regular email (Araki et al., 2006). However, guidance can also be provided orally; Bischof et al. (2008) provided guidance in the form of three to four 30 min telephone calls, while Kiluk et al. (2016) provided guidance in the form of 10 min face-to-face meetings at an addiction clinic over 8 weeks.

Although human guidance may enhance engagement with, and effects of, internet interventions, involving guides in the delivery of an intervention is obviously resource-demanding, particularly when the guidance constitutes lengthy unstructured interactions resembling a therapy session. Thus, to optimize cost-effectiveness of delivery, there is a need for research agendas to investigate how to best capitalize on the noted benefits of human guidance while minimizing time spent by the guide. In the current trial, we aimed to study the previously evaluated internet intervention Alcohol Help Center (AHC) (Cunningham, 2012; Cunningham et al., 2017) delivered with brief email guidance (i.e. semi-standardized emails during the first two weeks after gaining access to AHC). Our primary hypothesis was that participants allocated to AHC with brief email guidance would display significantly greater reductions in drinking at the 3- and 6-month follow ups compared to participants allocated to AHC with no brief guidance. We also hypothesized that participants allocated to the guided group would display greater engagement with the program in terms of a greater number of log-ins compared to the unguided group.

Materials and methods

Design

We conducted a parallel, two-group randomized, controlled trial with a 1:1 allocation ratio among internet help seekers with problematic alcohol use. The trial was prospectively registered at Clinical Trials (NCT03601793), and the Research Ethics Committee at the Center for Addiction and Mental Health in Toronto approved the trial (REB protocol #029-2018). Participants were recruited between July and November 2018.

Procedure

Advertisements were placed on Google Adwords and Facebook ads across all provinces and territories in Canada. The advertisement targeted people who were “experiencing difficulties in controlling or cutting down on their drinking” and who were interested in participating in a study “to help improve an online intervention for alcohol problems.” Potential participants could click on a link in the advertisement which led them to the study website where a brief description of the study was presented. Those who were interested in participating proceeded to the next page and completed an eligibility screening survey. Potential participants not found eligible in the screening survey were informed of their ineligibility and were provided a link to the public version of AHC. Those found eligible in the screening survey were asked to provide online informed consent and contact information (email address, telephone number and mailing address), as well as permission for study staff to contact them via phone or mail for follow-up surveys if correspondence by email was unsuccessful. Eligible participants then completed a baseline survey. Participants who completed the baseline survey were sent an email within a couple of days which contained a link to a study-specific web portal along with a unique log-in password to create an AHC account. Those who clicked on the link and used the password to create a new AHC account within a month of receiving the email were included in the study and randomized without stratification into one of two conditions: AHC with brief guidance or AHC without guidance. Randomization was carried out via an automated replicable algorithm setup by a member of the research team not involved in the day-to-day conduct of the trial. Participants were blinded and were not informed about study hypotheses. All participants were followed-up at 3 and 6 months post-randomization using an online survey (a notification was sent as a link to the participant’s email address). Participants completing the 3-month follow-up were sent a $10 Amazon.ca gift certificate and those completing the 6-month follow-up were sent an additional $10 Amazon.ca gift certificate (i.e. maximum total honorarium of $20 for each participant). Participants who did not respond to the initial prompt to complete each follow-up survey were sent up to three email reminders.

Participants

Inclusion criteria in the eligibility screener was being ≥18 years of age, having a score of ≥8 on the Alcohol Use Disorders Identification Test (AUDIT) and reporting a preceding week consumption of ≥14 standard drinks. Figure 1 shows participant flow throughout the trial.

Figure 1. Study flow diagram.

Main intervention: alcohol help center

AHC is a self-help website intended to be used repeatedly over an extended period of time. It contains 20 modules commonly used in CBT treatment of alcohol problems.

The modules of AHC are divided into three sections:

1. Getting started (10 modules focused on initiating change)

2. Dealing with difficulties (6 modules pertaining to key issues that often occur when working on change, such as dealing with drinking urges and managing relationships)

3. Maintenance (4 modules designed to help participants maintain their change).

In addition to the modules, a series of interactive tools are also available to help the participant in the change process, such as a drinking diary where the participant is encouraged to track his or her drinking, a blood alcohol calculator and an email and text messaging educational system that provides participants with encouragement and tips to deal with drinking concerns. AHC also offers access to a support group where users of AHC can post questions to other users and to health educators who monitor the support group (Cunningham et al., 2008). The health educators monitoring the support group were not part of the research team, and use of the support group among participants in the trial was not monitored by the researchers of this trial.

Conditions

Condition 1: alcohol help center with brief email guidance from a health educator

Those assigned to this condition were provided access to AHC and in addition to this received a series of four standardized emails sent out during the initial two weeks after randomization (see Appendix 1). The standardized emails were jointly developed by the authors, sometimes slightly individualized by the health educator (author CSu). The purpose of these emails was to present the structure of the intervention to the participant, guide the participant through the initial modules, boost confidence in change and promote future use of the modules and the AHC support group. In addition to the four standardized emails, the health educator could also respond directly to participant questions or comments. There was no set limit to the amount of guidance that the health educator could provide during the first two weeks. However, after the two weeks, the health educator did not respond to any more emails (the time boundaries surrounding the guidance was clearly articulated to participants in the first email).

Condition 2: alcohol help center

Those assigned to this condition were provided access to AHC but were not provided any guidance or any other form of contact with the research team.

Assessment and outcome measures

Participants completed assessments at screening, the 3 month follow-up and the 6 month follow-up.

Screening survey

Before consenting to participate in the trial, potential participants filled out a screening survey (AUDIT and number of standard drinks in preceding week) so as to be assessed for eligibility. The screening survey was followed by the consent form, after which participants were able to continue to the baseline survey.

Baseline survey

In the baseline survey, all participants were asked about demographics, i.e. age, gender, marital status, education, gross family income and employment status. In addition, participants were asked to respond “yes” or “no” to each of the 11 DSM-5 criteria for Alcohol Use Disorder (AUD) (American Psychiatric Association, 2013). Further, participants were asked whether they ever had received any form of treatment/help for alcohol problems and if so, what type. These questions about previous treatment/help were also administered at both follow-ups.

Primary outcome

At all three assessment points, participants were asked about the number of standard drinks (13.6 grams of ethanol) consumed during each of the preceding seven days. Two primary outcomes were calculated from this: 1) total number of standard drinks and 2) number of heavy drinking days (HDD) defined as ≥5 (men) or ≥4 (women) standard drinks on a single day.

Secondary outcomes

In addition to the preceding week alcohol consumption, the Alcohol Use Identification Test (AUDIT) (Saunders et al., 1993) was administered to participants. This instrument provides a 10-item multi-faceted assessment of alcohol problems including alcohol consumption, dependence symptoms and negative consequences of drinking. From the AUDIT, the first three questions focusing on alcohol consumption are often calculated separately as the AUDIT-C (where the “C” stands for “consumption”) (Higgins-Biddle & Babor, 2018), as was the case in this study. In the two follow-ups, the first three questions of the AUDIT that cover alcohol consumption over the past year, were adjusted to cover the last three months. Also, the Euroqol-5D (EQ-5D) (Herdman et al., 2011) was used to assess quality of life.

Statistical analysis

The study was powered to detect a small to medium differential effect size at 3-month follow-up in terms of standard drinks between intervention groups with power calculations based on results from previous individual trials that have shown guidance to have an effect on drinking outcomes (Blankers et al., 2011; Sundström et al., 2016). Based on previously conducted studies on the intervention, we expected 20% attrition at each follow-up. A sample size of 118 per group was thus required to achieve an 80% power with a two-sided 5% significance level. All participants randomized were included in the outcome analysis according to intention-to-treat principles. Using SPSS, multiple imputation on primary and secondary outcomes with five imputed datasets was performed. Generalized estimating equations (GEE) with an unstructured working correlation matrix and a 2 (Intervention) x 2 (3 month follow-up, 6 month follow-up) design was used to assess intervention effects on primary and secondary outcomes. The screening measure of the outcome analyzed was included as a covariate in each separate analysis. For primary outcomes (standard drinks preceding week and heavy drinking days preceding week), a negative binomial model was used due to the non-normal distribution of these variables, while a normal model was used for secondary outcomes. Planned two-sided contrast tests were used to test between-group differences at 3 and 6 months. Analyses were performed using SPSS 25 (IBM).

Results

Baseline characteristics

3185 individuals clicked on the link in the advertisement. Of these, 1348 (42%) individuals completed the eligibility screener and were considered eligible. Of these, 375 (28%) provided informed consent and 368 (27%) went on to finish the baseline survey and were sent the link. Of these, 238 (65%) used the password and created an AHC account and were thus included and randomized, with 118 randomized to the brief email guidance group and 120 randomized to the no guidance group. The 3-month follow-up rate was 56 (47%) in the brief email guidance group and 66 (55%) in the group without guidance, while the 6-month follow-up rate was 56 (47%) in both groups. There were no significant differences in age, gender, standard drinks, heavy drinking days, AUDIT or quality of life between follow-up completers and non-completers.

The mean age of the sample was 42.3 years and 65% were female (see Table 1). Participants reported a mean of 39.4 standard drinks and 4.6 heavy drinking days in the preceding week. As participants in the guidance group had significantly more heavy drinking days at screening (p = 0.013) and had visited Alcoholics Anonymous to a greater extent (p = 0.034; see Table 1), these two variables were entered as covariates in all outcome analyses.

Table 1. Screening and baseline variables

	AHC guidance (n = 118)	AHC no guidance (n = 120)	t/x	p
Mean (SD) age	42.9 (12.2)	41.6 (11.9)	−0.820	0.413
% Male	29.7	40.8	3.252	0.071
% Married/common-law	57.6	60.0	0.138	0.710
% Some post-secondary education	67.8	65.8	0.103	0.748
% Gross family income $50,000 or more	60.2	65.0	0.593	0.441
% Full-time/part-time employed	78.0	80.8	0.299	0.584
Positive AUD criteria (SD)	8.5 (2.1)	8.5 (2.1)	0.282	0.779
Drinks preceding week (SD)	40.8 (21.5)	38.0 (20.4)	−1.024	0.307
Heavy drinking days preceding week (SD)	4.9 (1.9)	4.3 (2.1)	−2.513	0.013*
AUDIT (SD)	25.1 (6.4)	25.3 (6.2)	0.202	0.840
AUDIT-C (SD)	9.9 (1.6)	9.9 (1.5)	0.266	0.791
EQ5D (SD)	0.74 (0.14)	0.76 (0.13)	1.472	0.142
EQ-5 VAS (SD)	56.9 (20.6)	57.8 (20.5)	0.355	0.723
Previous use of help for alcohol problems
% Have used online material	21.2	24.2	0.301	0.583
% Have used paper self-help material	32.2	25.0	1.513	0.219
% Have been patient at detox	9.3	6.7	0.571	0.450
% Have received outpatient treatment	11.0	13.3	0.298	0.585
% Have received inpatient treatment	5.9	5.8	0.001	0.974
% Have received counseling by a health-care professional	30.5	35	0.545	0.460
% Have visited Alcoholics Anonymous	33.1	20.8	4.517	0.034*
% Have received some other form of help	5.9	8.3	0.517	0.472

(^*= ≤ 0.05)

Guidance emails

The brief guidance group received a mean of 4.2 emails from the health educator (author CSu). The vast majority (84%) only received the four standardized emails that all participants in this condition received. Fifteen participants were sent 5 emails, two participants received 6 emails and two participants received 7 emails. Two participants asked to unsubscribe and one participant asked to withdraw from the trial before the four standardized emails had been sent out. In addition, one participant was mistakenly sent only three emails by the health educator.

Number of log-ins

Throughout the trial until the 6 month follow up, the number of times that participants logged in to AHC ranged from 1 to 61 times. Number of log-ins was categorized into four categories as the distribution of log-ins was skewed; 1 log-in = 45.8%, 2 log-ins = 16.8%, 3 log-ins = 12.2%, 4 or more log-ins = 25.2%). There were no significant differences in number of log-ins between the two groups (X²: 4.193, p = 0.241) (Table 2).

Table 2. Number of log-ins throughout the 6 month follow-up.

	AHC guidance	AHC no guidance	Total
1 log-in	51 (43%)	58 (48%)	109 (46%)
2log-ins	17 (14%)	23 (19%)	40 (17%)
3 log-ins	19 (16%)	10 (8%)	29 (12%)
4 or more log-ins	31 (26%)	29 (24%)	60 (25%)

Outcomes

Results showed no significant differences between the groups in any of the primary or secondary outcomes (see Table 3 for a presentation of estimated marginal means from the GEE model and t-tests of differences between groups).

Table 3. Primary and secondary outcomes at 3- and 6-month follow-ups.

	AHC guidance (n = 118)		AHC no guidance (n = 120)
	Mean	SE	Mean	SE	t	p
Drinks preceding week
3 month follow-up	20.4	1.3	20.3	1.2	0.028	0.977
6 month follow-up	19.0	1.1	20.3	1.1	0.809	0.419
Heavy drinking days
3 month follow-up	2.5	0.2	2.5	0.2	0.036	0.971
6 month follow-up	2.2	0.2	2.5	0.2	1.009	0.314
AUDIT
3 month follow-up	20.2	0.7	20.8	0.7	0.604	0.546
6 month follow-up	18.6	0.8	19.4	0.7	0.704	0.482
AUDIT-C
3 month follow-up	7.95	0.29	8.25	0.25	0.767	0.444
6 month follow-up	7.98	0.28	7.99	0.31	0.007	0.995
EQ5D
3 month follow-up	0.78	0.02	0.79	0.02	0.366	0.714
6 month follow-up	0.78	0.02	0.78	0.02	0.173	0.863
EQ5D VAS
3 month follow-up	64.7	1.9	66.0	1.9	0.483	0.630
6 month follow-up	64.4	2.8	64.8	2.6	0.102	0.919

Other help/treatment throughout the follow-up periods

The guidance group reported having used paper self-help materials significantly more often at the 3 month follow-up (p = 0.019). There were no other significant differences between the groups in extent of having sought treatment/help, either at the 3 month or 6 month follow-up.

Discussion

We found no evidence that providing brief email guidance resulted in greater drinking reductions compared to the group without brief guidance, and neither did we find any differences in secondary outcomes between the two groups.

Although we encouraged participants in the brief guidance group to respond to the emails sent to them by the health educator, only 16% of participants chose to do so, which was a surprisingly low number. For comparison, in a recent trial where guidance in the form of secure messaging over the intervention platform was offered to participants for five weeks, 33% of participants chose to engage in conversations with the guide (Boß et al., 2018). We don’t know why so few participants were willing to engage in conversation in the current trial. One possibility is that some participants did not realize that they had access to a health educator, but mistakenly assumed that the emails were automated or interpreted them as spam and consequently deleted them. As we had no verbal contact with participants at any point during the study, we were not able to verify to what extent participants in the brief guidance group understood that they had access to guidance from the health educator. Another possible explanation for the low rate of interaction with the health educator include email-related technical issues, such as participants not receiving the emails as intended or emails being automatically filtered as spam or junk mail and deleted. Compared to passively waiting for users to log in to a platform to access guidance, email provides a more direct method to reach out and connect with individuals. However, the possibility of technical issues point to specific limitations related to email as a guidance format. In addition, the low rate of interaction may also be due to the guidance being restricted to only two weeks. Perhaps this time period was simply too brief, in effect deterring participants from initiating a conversation with the health educator. Specifically, most previously published studies, where significant differences in favor of guidance have been found, provided guidance over longer period of time (Riper et al., 2018), and it is possible that there is a minimum time period that guidance needs to be offered in order to provide any added effects. Future trials could strive to identify differential benefits of guided interventions, and use factorial designs to simultaneously manipulate formats (i.e. email, secure messaging, telephone, face-to-face), length and intensity. Further, there is a substantial lack of qualitative research in the field. Although there are a few studies investigating experiences of users of unguided internet interventions for problematic alcohol use (Khadjesari et al., 2015; Marley et al., 2016), to our knowledge there are no studies specifically investigating experiences of users of guided interventions for problematic alcohol use. Such research could potentially aid in understanding what promotes interaction and what does not.

We found no evidence that brief email guidance increased engagement with the intervention in terms of number of log-ins. Even simple strategies such as automated emails (Titov et al., 2013), and more human-based strategies such as telephone support and facilitation of a discussion group (Carolan et al., 2017) have previously been found to increase engagement, but there was no evidence that email guidance increased engagement in the current trial. In addition, engagement was low overall, with almost half of participants logging in only once. It should be noted, however, that we did not investigate which AHC modules were accessed by participants, so we cannot rule out that module use differed between the groups.

Overall, participants displayed about a 50% reduction in terms of number of standard drinks and number of heavy drinking days when comparing the screening to the two follow-ups. As we did not include a no intervention control group in the study, it is unknown to what extent both interventions were effective in reducing drinking, whether reductions were due to assessment reactivity (Clifford & Maisto, 2000) or whether they were simply due to participants being highly motivated to reduce their drinking when applying to the trial. Interestingly, the large reductions in drinking observed at both follow-ups were not associated with an increase in quality of life. Instead, quality of life remained at virtually the same level from screening to follow-ups, suggesting that sharp reductions in alcohol consumption among people with problematic alcohol use are not necessarily related to an increase (or decrease) in quality of life. This finding is in line with results from a previously conducted study on an internet intervention for problem drinkers, where quality of life, also measured with EQ-5D, did not increase despite significant reductions in drinking (Essex et al., 2014).

Another observation is that participants recruited in this trial displayed more severe drinking problems than participants in previous trials evaluating this intervention (Cunningham, 2012; Cunningham et al., 2017), with AUDIT and AUDIT-C scores both being around 20% higher at screening in the current trial. That severity of drinking problems was high among participants was also reflected in the self-report data on DSM criteria, where almost 90% of participants had 6 or more positive criteria, indicating severe alcohol use disorder. A possible reason for this difference could be related to recruitment methodology. Participants in the previous trials were mainly recruited through newspaper ads. In the current study, however, participants were mainly recruited through search engine ads (meaning that participants were shown the study ad only when searching the Internet for terms related to drinking), and it may be that this form of recruitment reaches a more severely affected population. Lastly, our only significant finding between the two groups was that participants in the guidance group reported having used paper self-help material to a greater extent at the 3 month follow-up. Although we did not hypothesize that the email guidance would prompt further help-seeking, this finding is noteworthy. As this population is well known for procrastinating (Cunningham et al., 1993), future research could further explore whether email (automated or human guided) consistently increases help-seeking.

Limitations

Attrition was notably higher than in previous trials evaluating this intervention and should be mentioned as a serious limitation to the interpretation of the results. Although attrition was high in the trial, it did not differ between the two groups, which under the circumstances can be seen as a strength, since differential attrition is considered a major threat to internal validity (Crutzen et al., 2015). We used multiple imputation to deal with the attrition to be able to conduct an intention-to-treat analysis. A further limitation is that we did not monitor use of the AHC support group. In a previous study, around 30% of users posted in the support group (Cunningham et al., 2008). It is possible that the use of the AHC support group diluted any potential benefits of the brief guidance.

Conclusions

Brief email guidance from a health educator did not provide any added benefit in drinking outcomes or in engagement with an internet intervention aimed at problem drinkers.

Declaration of interest

The authors declare no conflict of interest.

Appendix 1

Scripts of the emails

Message 1

Hi, [name]!

I can see that you have successfully logged in to the Alcohol Help Center website. My name is [name]. I am a health educator and my job is to provide you with some initial assistance in starting to use this program. I will be sending you emails during the coming two weeks. You are welcome to respond if you like but after these initial two weeks I will not be able to respond to your emails.

Even though I don’t know you, I do know that you want to change something about your relation to alcohol (otherwise you wouldn’t have applied for this trial). I would first of all like to applaud you for taking that step. It is an important achievement in and of itself. To help you in achieving the goals that you might choose to set for yourself when it comes to alcohol, you now have access to the website Alcohol Help Center. Consider this website a valuable self-help toolbox that can be of great use for you in this process. Below, I will give you a brief introduction to the website and some important exercises.

Start by logging in to the Alcohol Help Center (website link) with your email address and password (let me know if you ever have trouble logging in).

On the welcome page, the first page that greets you after having logged in, take a look at the left hand side. There you will find three sections:

Getting started

Dealing with difficulties

Maintenance

Start by having a look at the ten exercises under the first section “Getting started” (click the + -sign and the exercises will appear). All of these exercises are valuable tools in the initial stages of change and deserve some of your time. However, I would particularly recommend that you in the coming days have a look at the following two:

Check your drinking. In this exercise you receive automated feedback based on your responses to a set of questionnaires about you and your relation to alcohol. This feedback is presented in the form of diagrams. The exercise also provides you with information about health consequences of alcohol.

The first two weeks. We know from research that the first two weeks can be the toughest and most challenging when initiating a change in alcohol habits. Therefore, it is important that you have effective strategies in place to counter challenges when they arise. In this exercise you are given nine very specific tips on how to succeed with this that many others have found helpful. Choose at least two of these tips to help you during the coming weeks, and make sure to write them down on a piece of paper or save them on the website. Remind yourself of these tips daily. As you can see, there is a host of additional exercises available for you in the “Getting started”-section such as Cutting Back and Costs and Benefits. Please, have a look at these as well.

After having done these exercises, it might be time for you to start formulating a drinking goal for the coming months. What is your ultimate goal in relation to alcohol? Do you want to stop drinking completely or do you want to cut down on your drinking? If you want to cut down, how much do you want to be able to drink? In what circumstances do you want to be able to drink? Keep in mind that the more specific you are with what you want, the easier it will be to follow these rules. And, again, it is always a good idea to write things down, whether you do it on the website or on a piece of paper.

I’ll briefly mention the interactive tools that you find right above the three sections: 1) Blood alcohol calculator were you can be provided with a rough estimate of how much alcohol you have in your body at any point in time. 2) Member goals where you can take part of inspirational goals set by other people who over the years have used this program successfully (read and let yourself be inspired!) and 3) Daily diary, where you can register your alcohol consumption day by day (beer, wine and spirits). Many people find it extremely helpful to keep a drinking diary, as it helps to identify patterns that you might not have been aware of. It is therefore highly recommended that you use this tool continually.

Lastly, there is also a support group where users of the website can ask questions and provide support for each other (click on Forums on the top left side). A health educator moderates the forum.

Continue to be active in this important process that you have initiated, and I will get back to you soon.

[name], health educator

P.S. If you do not wish to receive these emails, contact the research team

email: [study email]

In order to tailor the treatment to the individual participant, messages 2-4 will start with some reflections on the participant’s preceding response, about 2-3 sentences. If the participant has not responded, only the standardized message will be sent.

Message 2

Hi again,

(brief reflections on the participant’s response, 2-3 sentences. If the participant has sent no response, only standard message)

First of all, let me know if you are experiencing any problems with logging in to the website.

Hope you have found some interesting reading on the website in the last couple of days. If you haven’t had time, no worries. The important thing is that you are on track. Hope you found the tips in the exercise “The first two weeks” useful. Did you use any of the strategies? Also, did you decide on a personal drinking goal?

When you have decided to make a change but have not yet started to act upon it, it’s often helpful to understand the function of your behavior. This can be easier if you ask yourself questions like: In what situations do I generally drink? What is it really that makes me want to drink in these situations? The following two exercises are found in the “Getting started”-section, and both deal with mapping out these situations where you normally drink and understanding the function of them.

Triggers. In this exercise, you can identify situations in everyday life where you experience “triggers” to drink. You will find a list of triggers that are common among people with alcohol problems, but you are also free to list your own personal triggers that are not on the list. By identifying what it is that triggers you to drink, it may be easier for you to deal with these situations without drinking when they come about. Preparation is key.

High-risk situations. In this exercise, you will rate fifteen common high-risk situations for people who drink by the intensity you feel in them.

These exercises can help you to identify the situations that make you most vulnerable and that you should be most mindful of when encountering in the future. You will begin to circle both inner (feelings, thoughts) and outer (places, people) circumstances that are related to your drinking. That is an important part of changing your drinking habits.

Good luck with these exercises. Will get back to you soon.

[name], health educator

P.S. If you do not wish to receive these emails, contact the research team

email: [study email] or telephone [study telephone number]

Message 3

Hi again,

(brief reflections on the participant’s response, 2-3 sentences. If no response, only standard message)

You have now been an active user of the program for over a week. Hope you had use of the exercises Triggers and Risk situations. Did you become aware of triggers that you hadn’t noticed before? Was it easy to identify which high-risk situations that you experience most intensely?

The next thing I would like for you to have a look at are the six exercises in the next section “Dealing with difficulties”. The exercises in this section are relevant for anyone who has 1) decided what kind of change it is that they desire (in terms of drinking goals) and 2) started mapping out their risk situations. You need to prepare yourself for risk situations that will arise in the future, so that you can deal with these situations and still keep to your drinking goal. I would particularly like you to have a look at the following two exercises.

Dealing with desires. Feelings of desire and urges are normal after you have quit or reduced your drinking. Therefore it is important to have a list of activities ready that can help you when the desires or urges set in. In this exercise you can choose from a list of commonly used methods of resisting desires or urges for alcohol, as well as add your own personal methods should you wish. It’s always best to be well-prepared!

Saying no thanks Alcohol is often consumed in social situations. If you are a person who often drinks alcohol with other people, you probably know that it can be very hard to say no to a drink once you are in the situation. Often, people around you expect you to drink, since you always have before. In this exercise, you get help in preparing yourself to say no to alcohol in social situations.

Good luck with these exercises. I will get back to you soon.

[name], health educator

P.S. If you do not wish to receive these emails, contact the research team

email: [study email] or telephone [study telephone number]

Message 4

(brief reflections on the participant’s response, 2-3 sentences. If no response, only standard message)

Hi again,

This is the last message from me. I hope that you have gotten some new insights these past two weeks and I strongly encourage you to keep working with all of the exercises over the coming months.

The last section “Maintenance” contains four exercises that will be important for you in the coming weeks. I would suggest you have a look at these four exercises in the coming days, they are all important. In the Support team exercise, you will be asked to set up a team (family and/or friends) that can provide support for you when you experience difficulties. Building a support team can make a tremendous difference. You will know who you can contact when you are faced with a difficult situation. Another important exercise is Keeping active, where you can make a list of things to do to keep you physically active. Keeping physically active is important, and can both help you to feel better and help you to deal with desires and urges. Connected to the importance of feeling healthy is the exercise Healthy eating. Reducing drinking often increases your appetite. It is important that you are aware of and prepared for that change, so that the reduction of alcohol doesn’t result in a major weight increase. Finally, stress is an important risk factor and in the exercise Coping with Stress you can learn a lot about strategies that help you deal with stress in a constructive way.

Alcohol Help Center is a rich website, and there is probably still a lot left for you to explore. I encourage you to keep working with the website and make sure to find a way to use it regularly. Also, just because you have done an exercise, it doesn’t mean that you can’t do it again. Often, when you redo these exercises, you discover a lot of new things.

It has been a pleasure guiding you through these past two weeks. We will get back to you in about 10 weeks to ask you to fill out some questionnaires. In the meantime, I wish you the best of luck with your ambition of changing your relation to alcohol!

[name], health educator