Internet Support for Dealing with Problematic Alcohol Use: A Survey of the Soberistas Online Community.

Introduction

Research into online groups and social networking sites for the management of a variety of health conditions suggests the positive role these groups play in offering support in a way that might not be available in traditional treatment services, in addition to the convenience and anonymity offered by a virtual meeting place (e.g. Greene ; Jones ; Christie, 2013; Stewart-Loane and D'Alessandro, 2013; Chung, 2014; Lockhart ; Stewart-Loane ).

The stigma attached to alcohol use disorders (AUD) results in lower levels of disclosure about alcohol use and acts as a barrier to seeking treatment (Jones ; Probst ), especially in treatment-naive people who may be uncertain how to construe their difficulties, and whether and where to seek appropriate help (Schuler ). With advances in technology, there are increased possibilities with the Internet, and other forms of social media for people with AUD to seek help (Cunningham ; Hester ). This allows those who are unwilling or unable to go to in-person services to access support (Vernon, 2010; Hester ) and may also facilitate longer term self-management by people with AUD in a similar way to other long-term conditions (McKay and Hiller-Sturmhöfel, 2011).

There is now substantial evidence of the effectiveness of online interventions to reduce alcohol use in a range of increased risk drinkers drawing on Cognitive Behavioural Therapy and Self-Management and Recovery Training (SMART) principles (see White ; Riper ; Hester , Riper ). However, there is limited addiction-specific research exploring the use of online groups that are underpinned by mutual aid and social support; the few studies that have considered online support groups (e.g. Humphreys and Klaw, 2001; Cunningham ; Coulson, 2014) agree that the Internet plays a crucial role in overcoming some of the physical and emotional barriers to accessing in-person support. Sharing success stories, helpful strategies and discussion of difficulties (Cunningham ), and disclosure of personal information, offering support and advice, and a shared goal of sobriety (Coulson, 2014) were identified as factors engaging people with online support. In a review of self-help groups more broadly, social support was found to be one of the ‘key ingredients’ (Moos, 2008), along with goal direction, structure, abstinence-oriented norms and role models. However, most of this research was within the 12-step community in the USA; there is still much to learn about the processes involved in online compared with in-person support.

The pace of change in the technology of social media and the increased expectations of those who use them provide a challenge to those designing platforms that optimise engagement and retention within online communities. But a review of mechanisms by which social media may have its effects in the self-management of chronic conditions considered this in terms of ‘Affordance Theory’ (Merolli ), i.e. when people interact with social media it is because of what they perceive it ‘affords’ them, concluding that the specific features and functionality of a social network platform are secondary to what they afford participants in terms of social interactions, information sharing, flexibility of access, more personalised support and an opportunity to share narratives (Merolli ).

The aim of this paper is to describe the population, component parts of a new peer-led online support group, ‘Soberistas’, and consider what the different parts of the site affords the members and browsers who use it, both as a form of mutual aid, but also to consider factors involved in designing an effective online platform for social support. Soberistas is a ‘social network site for people who are trying to resolve their problematic drinking patterns’ (Rocca, 2016). It was launched in 2013 by Lucy Rocca, 18 months after she became abstinent from alcohol, based on what she reflected had been her own needs at that time. The ethos is described as ‘non-prescriptive, non-religious, and non-judgmental’. This is the first study to survey the membership and to consider its place in online support, as well as to generate hypotheses about mechanism of action of such sites for future research.

Materials and Methods

A cross-sectional survey was designed comprising free text and fixed response questions, divided into four sections:

Demographics

Soberistas membership—Participants were either ‘browsers’ or ‘members’. Browsers have access to most content, but cannot contribute to discussions; subscription-paying members have access to all areas of the site and full functionality. Participants were asked their reasons for joining and continued use, how participants became aware of the website, use of previous support (e.g. Alcoholics Anonymous (AA), SMART, none).

Soberistas usage—Time spent on the site, which parts of the site do participants find most helpful and use most often. The website has five core features:

Personal stories—described as ‘optimistic and honest accounts of how we managed to get our lives back on track by ditching alcohol, … to help others who are looking for a way out of the booze trap’. These are submitted by members, can be viewed by anyone.

Various blogs and forums—these include lifestyle articles written by the founder and editor of the website, as well as posts written by members, which are available to all, and a discussion forum on a range of themes (e.g. from campaigning to how to beat cravings) which can only be contributed to by members.

Ask the Doctor page—members are able to submit questions and view all responses which are replied to monthly.

Webinars—given by a range of experts in health, lifestyle, etc., only available to members.

Chat room—offering support and discussion threads in real time, for members only. In addition, there are also sections which feature alcohol-related news stories, signposting to other resources, adverts for books and therapies, and features such as a ‘member of the month’ and a book club.

Alcohol consumption—Estimated length of problematic use, current/recent alcohol use and the impact of alcohol on five domains (physical health, mental health, close relationships, work/study and finances) using a six-point Likert scale. We also asked how consumption had changed since using Soberistas.

The survey took ~5 min to complete. It was piloted by 10 users of Soberistas, and minor adjustments were made following feedback about usability. Soberistas staff also reviewed and piloted the questionnaire but were not otherwise involved in its development, or the analysis of results. The study received a favourable ethical opinion (ERGO number 16245) and an invitation to participate in the survey was posted on the Soberistas website for 1 month.

Data from the online survey were imported directly into SPSS (SPSS Inc, 2015). Simple descriptive, and non-parametric statistics were used to analyse the numerical data and free-text responses were coded and categorised by S.C. and uncertainties discussed within the research team.

Results

Overview

At the time of the survey (August 2015), there were 32,550 registered users (all who had registered since the site was launched in 2013). Of these, there were ~3800 active users; 1828 were subscription-paying members and ~2000 active browsers.

Of the 1828 members and 2000 browsers, 438 participants (11.4%) completed the online survey, but 6 participants did not complete any of the alcohol data questions and so these were excluded from the analysis. The demographics of the sample are shown in Table 1.

Table 1.

Demographics of survey responders (n = 432)

Variable	n (%)
Gender
Female	404 (94)
Age (years)
18–24	3 (0.7)
25–34	16 (3.7)
35–44	112 (25.9)
45–54	181 (41.9)
55–64	89 (20.6)
64+	31 (7.2)
Ethnicity
Caucasian	419 (97)
IP address location
UK	311 (71.9)
USA	56 (12.9)
Australia	9 (2.1)
Canada	9 (2.1)
Other	47 (10.8)
Household composition
Lives alone	58 (13.4)
Lives with child(ren)	46 (10.6)
Lives with partner	141 (32.6)
Lives with partner and child(ren)	170 (39.4)
Other	16 (3.7)
Missing	1 (0.2)
Occupation
Employed	317 (73.4)
Of which self-employed	104 (24.1)
Retired	41 (9.5)
Homemakers	45 (10.4)
Other[a]	27 (6.3)
Missing	2 (0.5)
Highest qualification
Postgraduate degree	119 (27.5)
Bachelor's degree	142 (32.9)
School leaving	88 (20.4)
Further training qualification	38 (8.8)
None	17 (4)
Other	27 (6.3)
Missing	1 (0.2)

aOther includes student, long-term sick/disabled, maternity leave, etc.

Of those who completed the survey, 280 (64.8%) were fee paying members and 150 (34.7%) were browsers. As can be seen from Table 1, the vast majority of survey responders were white females, based in the UK. Half were living with children, either alone or with a partner, 73.4% were employed and only 1.5% (N = 5) describing themselves as long-term sick or disabled. Almost 70% had some form of post school-leaving qualification and 27.5% describing themselves as having a higher (postgraduate) degree.

A comparison of fee paying members with browsers showed that members were more likely to report being <1 year sober (86.4% vs 73.8%; X2 10.5; P = 0.001), but there was no difference as to whether they had sought help from any other source previously (64.3% vs 66.0%; X2 0.129; P = 0.720) or reported having an alcohol problem for >10 years (67.7% vs 61.0%; X2 1.95; P = 0.163).

Alcohol history

Information related to participants' reported alcohol use and access to previous support is shown in Table 2. Over 60% of participants rated themselves as having had problematic alcohol use for over 10 years (the longest response category available) and another 28.9% acknowledging problematic use for between 3 and 10 years. Almost half (46.5%) had not accessed any previous support, 28.9% had tried AA and <20% had asked or received help from a healthcare professional or service. Of those who had previously sought help, there was no difference in reported length of alcohol problem; 56.5% (>10 years) vs 49.1% (<10 years), X2 2.24; P = 0.134.

Table 2.

Alcohol use and previous support (n = 432)

Variable	n (%)
Length of problematic alcohol use
<1 year	13 (3)
1–3 years	22 (5)
3–10 years	125 (28.9)
10+ years	270 (62.5)
Missing	2 (0.5)
Previous support tried
AA	125 (28.9)
SMART recovery	29 (5.8)
Other online	50 (11.6)
Healthcare	83 (19.2)
Other[a]	72 (16.7)
None	201 (46.5)
Last drink of alcohol
Within 24 h	108 (25)
Between 1 and 7 days	77 (17.8)
>7 days <1 month	43 (10)
>1 month <6 months	87 (20.1)
>6 months <1 year	38 (8.8)
Over a year ago	77 (17.8)
Missing	2 (0.5)
Units consumed in the previous 7 days
0	238 (55.1)
1–35	111 (25.7)
36+	78 (18.1)
Missing	5 (1.2)
Change in alcohol consumption since joining[b]
No change (still drinking)	54 (12.5)
Now alcohol free	148 (34.3)
Maintained abstinence	47 (10.9)
Reduced consumption	103 (23.8)
Other[c]	79 ( 18.3)
Missing	1 (0.2)

aOther includes alternative therapies (e.g. hypnosis), private counselling, inpatient rehabilitation.

bBased on categorisation of free-text responses.

cOther examples: ‘more aware of the problem’, ‘too new [to the site]’ or just said ‘yes’.

Participants had clearly joined the website at different stages of their change in alcohol use; 10.3% had joined the site already abstinent, looking for support in remaining so; 34.3% described themselves as becoming ‘alcohol free’ since joining the site; 23.8% described themselves as having reduced their alcohol consumption and 12.5% with no change in their drinking behaviour.

Those who reported sobriety of longer than 1 year were more likely to have sought some form of help previously (21.7% vs 13.5%, X2 4.9; P = 0.026), and if they were fee paying members, were more likely to have joined over a year ago (27.0% vs 9.4%, X2 12.8; P = 0.001).

Impact of alcohol on various domains

All participants 52.7% (N = 228) who admitted to drinking alcohol in the last month were asked to rate the impact of alcohol in the last month on each of the following domains: physical health, mental health, close relationships, work/study and finances; using a six-point Likert scale to answer five questions: ‘In the last month, how much would you say your (e.g. physical health) has been affected by your alcohol consumption?’ (see Fig. 1).

Fig. 1.

Impact of alcohol on those still drinking over the previous month (n = 228).

Mental health was most likely to be rated as being ‘extremely’ affected by current alcohol use 112/219 (51%), followed by physical health 75/220 (34%) in those who had drunk alcohol in the last month. Close relationships, work/study, and finances were less likely to be rated as being ‘extremely’ affected, but still were so by 26%, 26% and 24% of participants respectively. In terms of finance 97/216 (44.9%) of participants rated them ‘not at all’ affected by current alcohol use.

Website structure and use

All aspects of the website were listed, and participants asked to rate how often they used each section and how helpful they found it.

Personal stories—these are submitted by members, can be viewed by anyone and were cited as ‘particularly helpful’ by 80.8%, and used sometimes/frequently by 91% of respondents.

Various blogs and forums—which are available to all, and a discussion forum on a range of themes (e.g. from campaigning to how to beat cravings) which can only be contributed to by members. These areas were cited as ‘particularly helpful’ by 73.1%, and used sometimes/frequently by 82% of respondents.

Ask the Doctor page—rated as ‘particularly helpful’ by 29.3%, and used sometimes/frequently by 59.6% of subscription-paying respondents.

Expert Webinars—rated as ‘particularly helpful’ by 34.6%, and used sometimes/frequently by 61.8% of subscription-paying respondents.

Chat room—rated as ‘particularly helpful’ by 19.6%, and used sometimes/frequently by 41.1% of subscription-paying respondents.

The majority of respondents (81%) spent between 1 and 3 h on the site, 13.9% spent between 4 and 10 h, and a 4.4% over 10 h on the site in the last week, and this was rated as ‘usual’ by 71.3%.

Reasons for browsing/joining the website

Responses to this free-text question were coded and summarised in Table 3. Each participant could list more than one reason.

Table 3.

Categorisation of free-text data: reasons for joining (n = 432)

Categorised response	n (%)	Example responses
Alcohol-related goals	280 (64.8)	‘To try and stop drinking alcohol.’ ‘To make me aware of how much I'm drinking.’ ‘Concern over alcohol consumption.’ ‘To help me stay sober.’
Support from others	130 (30.1)	‘Love the camaraderie—and I think many women like myself.’ ‘Like minded support.’ ‘To have people who understand to communicate with. Advice.’ ‘I'd given up drinking 18 months previously, had moved and wanted a support network.’
Curiosity in the concept/alternative form of support	46 (10.6)	‘I wanted to stop drinking and didn't want to go to AA.’ ‘I was looking for online support other than AA.’ ‘I felt it was very accepting and other forms of help available were stigmatising…I wasn't really sure if I was “bad” enough to need other alcohol services.’ ‘Soberistas seems strong, and I love the “normalization” of the problem.’
Media	61 (14.1)	‘Read about it in a newspaper article then signed up.’ ‘An article of Lucy's in Good Housekeeping.’ ‘Heard about them on a Radio 4 programme.’
Other reasons, e.g. Anonymity Recommendation Generic answers Information	38 (8.8)	‘My mum found this site and recommended it to me as it sounded just like my behaviour.’ ‘Self-awareness.’ ‘Good source of information.’

As can be seen from Table 3, almost 65% of responses expressed some goal related to changing their relationship with alcohol. These ranged from people wishing to increase their awareness of the problem to those who had already stopped drinking but were looking for support to help maintain abstinence. Support from a peer group with similar experiences was cited by 30% for their use of the site, while 10.6% stated they were exploring alternative options or were unsure if they would be suitable for other formal or peer-group support. Hearing or reading about the site opportunistically and finding a resonance with it was mentioned by 14% of respondents.

For the 280 participants who were paid-up members of the site, the reasons why they continued varied (see Table 4). As well as the personal goals of attaining abstinence or remaining alcohol free, respondents also cited the importance of being part of a community, and wishing to give something back to the site either financially or by peer support to those still struggling. Respondents also appeared to value it as a repository of useful and interesting information, not only around alcohol but on health and well-being which are part of the Soberistas philosophy.

Table 4.

Categorisation of free-text data: reasons for continuing membership of the site (n = 280)

Categorised response	n (%)	Example responses
Community support	151 (55.9)	‘It's good to have ongoing support, and to feel part of a community, and that I am not alone and a “failure”. Hearing about others struggling with this problem is helpful.’ ‘The sense of community with people who understand.’ ‘The help and friendships I've made.’ ‘Feel part of a family.’
Specific features of the site including information and advice	70 (25.9)	‘Being able to view webinars.’ ‘I find the blogs and notes encouraging.’ ‘I like to read other people's stories.’ ‘Belonging to an organisation that gives excellent advice to those abusing alcohol.’ ‘Interesting articles.’ ‘Get a lot of benefit with a wide range of issues not just alcohol.’
Alcohol-related goals	54 (20)	‘I will be sober 6 months in a week. Could not have done it without Soberistas and need Soberistas every day.’ ‘To remind me why I stopped drinking.’ ‘I don't think I would still be sober without the site.’ ‘I don't want to relapse.’ ‘Want to stop drinking.’ ‘I am still worried about my drinking.’
Wishing to give back (to the site as a whole, and to other members)	18 (6.7)	‘To support the site, and because it helped me so much.’ ‘Being able to contribute financially so that the site continues to be available for those that need it.’ ‘To be able to help others.’
Other; e.g. General interest Curiosity Cost Generic comments Convenience	43 (15.9)	‘Interest in the subject.’ ‘The membership fee is affordable.’ ‘Direct debit! Only look occasionally now.’ ‘Knowing it's in the background.’ ‘Being able to drop in as I wish.’
Will not be continuing	12 (4.4)	‘I won't be continuing it.’ ‘Not sure I will.’

Discussion

Although the survey was only completed by 15% of subscription-paying members and ~8% of browsers, the results give an indication of the users and processes of this online community. Most striking is that although almost two-thirds of the sample describe problematic alcohol use for over 10 years almost half had never tried any previous form of support. Of those still drinking, just over 50% describe the impact of alcohol on their mental health as being ‘extremely severe’. Combined with their reports that almost half were living at home with children there may well be effects on the health and well-being of a family. Respondents were primarily female, overwhelmingly in employment and with postgraduate qualifications. This resonates with findings from the US-based ‘Moderation Management’ (MM), an online platform for non-dependent drinkers (Humphreys and Klaw, 2001). Those who used MM only were significantly more likely to be female and have greater severity of AUD than those attending in-person meetings, and those with any online MM contact had higher levels of education. As with our study, the authors argued that online MM tapped into a specific group within the population who might otherwise not access alcohol-related services, and for whom the convenience and anonymity of online support networks are particularly beneficial.

Participants in our study describe coming to the site at various stages of change of their alcohol use, with 10% joining the site having already achieved abstinence. A similar number were still drinking but using the site for information and to explore what options were available to them, as well as to compare their situation with that of others. Stewart-Loane , 2014) studying online health platforms in general found that members with other chronic health conditions may alter their use of online communities over time depending on their needs—for example, at the start of membership their motivation may be to seek information, but as time passes and ongoing communication between other members enhances the value of the community, they start to create value for others by offering support or information, and so the design of the site, enabling people to make these transitions within a single platform without having to find a new site potentially breaking off helpful relationships, makes for ‘affordance’ (Merolli ).

In their review, Merolli suggested five categories that help to explore the underlying processes which may be involved for people joining virtual mutual aid groups, such as Soberistas; they termed these identity, flexibility, structure, narration and adaptation.

Identity and narration

People using condition-specific social media sites are afforded ‘more choice and control over how they present and assert themselves’ with features such as blogs and chat rooms allowing discussion of taboo or difficult topics more honestly than they might face to face, especially for stigmatised conditions. This is likely to be particularly important at the beginning of engagement with a site. Soberistas participants valued being able to read ‘personal stories’ of people they identified with, responding to posts and using blogs and forums.

Flexibility

AA has frequent meetings in many geographic locations such that an individual can regulate how many meetings they attend based on their own needs. An online community facilitates a similar self-regulation of need by being constantly available. The review suggested that ‘asynchronous’ communication (i.e. people being able to post comments at any time, and are not required to reply immediately, unlike chat rooms, or face-to-face meetings) may aid rather than hinder communication and enable people to engage with topics and emotions when they feel ready to do so. The literature for people with addictions and problematic drinking (Finn, 1996; Humphreys and Klaw, 2001; Cooper, 2004) suggests that the convenience of an online platform is helpful for some, and in this study, the range of time spent on the site (from <1 h to over 10 in the last week) suggests flexibility is important. However, the role of asynchronous communication as a facilitator for social support is something that requires further exploration.

Structure and adaptation

This refers to the ‘architecture of participation’—the different levels at which people connect with each other, share relevant information and facilitate self-management. This requires a range of functions (blogs, webinars, chat rooms, etc.) within the one site so as to enable participants to flexibly navigate between them according to their needs. In our study, participants reported different patterns of use of the site, which may be based on their stage of change and needs over time, and fits with findings of online supportive communities for other disorders (Stewart-Loane and D'Alessandro,2013; Chung, 2014; Stewart-Loane ).

Respondents to the Soberistas survey made reference, via free-text comments (see Tables 3 and 4), to a variety of social support behaviours enacted on the site, and this was a commonly cited reason for spending time on the site, and continued membership. Recent work describing a Social Identity Model of Cessation Maintenance (SIMCM, Frings and Albery, 2015) and a Social Identity Model of Recovery (SIMOR, Best ) highlights the importance of social identity processes in recovery; the authors argue that connection to recovery-orientated groups helps facilitate the development of a non-drinking social identity necessary for sustained recovery (although SIMCM takes a social cognitive perspective while SIMOR views social identity transitions within a changing social context from a systemic, rather than an individual, perspective). Both these models have relevance for how individuals engage with online peer-led communities, but in-depth empirical research is required to elucidate this further, including further qualitative work around the role of identity in the recovery process.

Community surveys have long shown that many people with alcohol-related problems can resolve these without formal treatment and those who make it into treatment services are often at the more severe end of the spectrum with a trajectory of a chronic relapsing and remitting illness. A recent review (White ) suggests that the ability to resolve alcohol problems depends on an interaction between different levels of personal vulnerability, severity, as well as individual and community recovery capital. At present, it is not possible to say whether the members and browsers of the Soberistas website belong to a group who would have a high percentage of natural resolution, due to higher levels of social capital (Granfield and Cloud, 2001), or whether it offers an earlier intervention to people before they reach the stage of a more severe, chronic and relapsing condition with the associated loss of their protective factors of work and family. However, given that the majority in this study describe long-standing problematic use, and around half had sought other forms of assistance, with those who reported sobriety of longer than a year more likely to have sort help previously, it may be that Soberistas is offering this group something different than that previously available within the treatment system.

The results are limited by the relatively small sample size, the cross-sectional nature of the survey and the inability to validate the veracity of any online responses. However, they do suggest that the site provides a supportive online environment to a group which has not successfully engaged with treatment or other forms of peer support. As technological advances make online social support increasingly accessible, the opportunity to develop effective peer support online will require an understanding of the needs of specific groups, in terms of models of social support, but equally how best to engage specific populations and retain participation in online communities, as well as developing appropriate research methods to capture this interaction.