Online support needs and experiences of family members affected by problem gambling

Research Report – April 2017

Approach

Participants and recruitment

Recruitment of family members occurred between January 2014 and January 2015 from Gambling Help Online. Criteria for participant inclusion was (1) a family member or friend of someone with a gambling problem; (2) completion of an e-therapy option, including chat, email, forum or website information; (3) an interest in taking part in a research project; and (4) a willingness to be contacted following engagement with an e-therapy option.

An email was sent to 407 eligible family members who met the inclusion criteria. A total of 78 family members clicked through to the survey. Consent was sought from 65 family members meeting the inclusion criteria and three declined to participate in the study. The final sample comprised 62 family members. The survey was offered entirely online via Qualtrics survey software.

Ethics approval for this study was received from the Eastern Health Research and Ethics Committee E01/2014.

E-therapy options

Gambling Help Online provides chat, email, website, forums and self-help. Family members accessing any of these services were included in the current report. Gambling Help Online is a nationally funded program that is operated by Turning Point from the state of Victoria.

E-therapy services are offered 24 hours a day, seven days a week. All services are free of charge and without appointment. Professional counsellors with backgrounds in psychology and social work deliver e-therapy. The specific options investigated in the current study:

  • Synchronous chat: Chat is offered 24/7 and works similarly to instant messaging, where both the counsellor and client type in a secure environment. A typical counselling session has a 45-minute duration.
  • Asynchronous email: Email support is provided via the same secure site as the real time chat. A client is allocated the same counsellor for two to three emails a week for approximately six weeks.
  • Website: The website provides information on gambling issues, interactive self-assessments, and strategies for regaining control as well as accessing support and helping others. In total, the site offers over 30,000 words of content across more than 20 separate pages.
  • Community forums: Forums are post-moderated by a clinician from Gambling Help Online. Family and friends can read and create a post in the forums in a family specific area or in general topics such as strategies for change and stories of recovery. Critically, this service provides peer support from a supportive community of members.

Measures

The survey included very brief demographics, measures of the impact of gambling and services accessed. Survey items were initially developed by the lead researcher and then subjected to consensus amongst the research team and AGRC. The survey was then pilot-tested with counsellors, gambling researchers and administration staff.

Characteristics of family members

Family members reported their gender, age and an email address. In addition, we asked family members to state the nature of the relationship with the person with the gambling problem (i.e., "What relationship do you have to the person with the gambling problem?"). The response options were partner, parent, sibling, child, friend and other relative. Given this broad range of relationships, family members were advised that when the questionnaire referred to family, it was inclusive of friends and colleagues as part of that family.

Family members were asked to state whether they currently lived with the person with the problem (i.e., "Have you lived with your family member with the gambling problem in the last 12 months?"). To determine the impact of gambling on the family member, we asked family members to provide information on the time since the problem was identified by the family member. Specifically, we asked, "To the best of your knowledge, for how long has your family member (the gambler) had a problem with gambling?", and then, "How long has it been since you first discovered that your family member (the gambler) has a gambling problem?"

Impact of gambling on family members

Gambling impacts and coping were measured with multiple scales. These were:

  • The Problem Gambling Significant Other Impact Scale (PG-SOIS) (Dowling et al., 2014) is a six-item scale that is a brief measure of the impact of problem gambling across six broad domains. These include financial, emotional distress, interpersonal relationship with the gambler, social life, employment and physical health. The PG-SOIS measures the impact in the previous three months using a frequency response format from (0) "Not at all" to (3) "Often", with scores ranging from 0 to 18. This is the first brief scale that seeks to measure the impact of gambling on family members but it has not yet been subject to validation.
  • The Problem Gambling Family Impact Measure (PG-FIM) (Dowling et al., 2015) is a 14-item scale that measures three domains of functioning. These include financial, increased responsibility and inter/intrapersonal functioning. Compared with the PG-SOIS, the PG-FIM investigates fewer domains but does so in more depth. The PG-FIM measures the impact over the previous three months using a frequency response of (0) "Never" to (3) "Often", with results in a possible score ranging from 0 to 42. The PG-FIM also has three sub-scales including financial impacts (items 1, 2 and 3), increased responsibility impacts (items 4, 5 and 10) and psychosocial impacts (items 6, 7, 8, 9, 11, 12, 13, 14). The means of the factor scores indicated that the most commonly endorsed factor was psychosocial impacts (M = 2.50, SD = 1.47), followed by financial impacts (M = 2.09, SD = 1.10) and increased responsibility (M = 1.39, SD = 0.81). The PG-FIM (problem gambler version) has displayed good internal consistency: financial impacts (α = 0.87), increased responsibility impacts (α = 0.85), and psychosocial impacts (α = 0.94) but still needs to be validated against other measures of family impact (Dowling et al., 2015).
  • The Coping Questionnaire (Orford, Templeton, Velleman, & Copello, 2005) measures strategies family members use to cope with alcohol use problems. Adapted for gambling from a similar screen for family members, the 30-item instrument has been used to measure coping in family members of people with problem gambling (Krishnan & Orford, 2002; Dowling, Suomi, Jackson, & Orford, in preparation). Dowling et al. (in preparation) identified six factors when this instrument was employed to measure the perception of family member coping from the perspective of treatment-seeking gamblers: Engaged-supportive coping (α = 0.88), involving engaging in trying to change a family member's excessive gambling in a variety of supportive and assertive ways; Engaged-emotional coping (α = 0.85), involving engaging in trying to change a family member's excessive gambling in a variety of emotional and controlling ways; Withdrawal coping (α = 0.82), involving withdrawing from the family member or engaging in activities independently of the family member; Tolerant-sacrificing coping (α = 0.74), involving putting up with a family member's gambling in a sacrificing way; Tolerant-emotional coping (α = 0.74), involving putting up with a family member's gambling in an emotional way; and Tolerant-accepting coping (α = 0.47), involving accepting the family member's gambling. The total summed score for all 30 items yielded high internal reliability (α = 0.93). Based on the highest loading item for each of these factors, we adapted the questionnaire so that it could be administered as a brief six-item questionnaire. Participants rated the frequency of use of each action over the past three months. The response format was (0) "Not used" to (3) "Often", with a range of scores between 0 and 18.
  • The Oslo 3-item Social Support scale was used to determine the level of social support available to family members (Dalgard, 1996). Questions included (1) "How many people are so close to you that you can count on them if you have serious problems?" (none, 1-2, 3-5, 5+); (2) "How much concern do people show in what you are doing?" (1 = a lot of concern or interest, 5 = no concern or interest); and (3) "How easy is it for you to get practical help from neighbours if you need it?" (1 = very easy, 5 = very difficult; note this item was reversed scored). Higher scores indicate greater social support than lower scores. Previous research has reported the OSLO is related to quality of life (Schmidt, Mühlan, & Power, 2006) and that the items can reliably be combined to measure a global score for social support (Kamenov et al., 2016).

Types of low- and high-intensity services accessed

We sought to identify a range of common services and supports that people commonly used to change or manage a family member's gambling problem. Responses were grouped into three categories: low-intensity, high-intensity and self-directed actions. Low-intensity interventions were defined as those offered as a single session and do not necessarily involve an ongoing commitment. Each of the options identified were provided at a distance and via the Internet (i.e., chat and email) or phone (i.e., helpline). Ongoing interventions (e.g., problem gambling counselling and financial counselling; support groups), as well as treatment from a general practitioner (GP), psychologist or psychiatrist were defined as high-intensity interventions. We also included a category of self-directed interventions (e.g., read information on websites; tried a self-help strategy like budgeting; talked to family members or friends about gambling).

Low-intensity interventions:

  • talked to a gambling help counsellor online;
  • sent an email to a gambling help counsellor;
  • phoned a gambling helpline.

High-intensity interventions:

  • talked to a gambling counsellor face-to-face;
  • sought financial counselling by phone or face-to-face
  • talked to a psychologist, psychiatrist or GP about the gambling;
  • attended a family and friends support group for gambling;

Self-directed interventions:

  • read or posted in the online forums;
  • read information on the Gambling Help Online website;
  • talked to family members or friends about the gambling;
  • tried a self-help strategy like budgeting to reduce the impact of gambling.

Family members were asked, "How often have you accessed support? Here is a list of services and supports people impacted by gambling can access. They include online options as well as face-to-face and phone services. This time we are checking whether you have ever used the service. How often have you ever tried these support options? If you have never used the service or support check 'never'." Family members responded with "never", "once", "2-5 times" or "more than 5 times".

Types of e-therapy accessed

Participants were asked three questions about their experiences with Gambling Help Online. Questions were based on an evaluation of the Michigan Gambling Helpline (Ledgerwood, Wiedemann, Moore, & Arfken, 2011) and included:

  • Thinking back to when you registered for Gambling Help Online yesterday, what service/s did you access? (Response options were "yes" or "no" against each service offered.)
  • Thinking back to the services you accessed yesterday, rate the extent to which the service you received was enough (i.e., that you did not need any more services for your concern)? (Response options were from 1 = "definitely not enough" to 5 = "definitely enough" for each e-therapy service accessed.)
  • Thinking about the service you accessed yesterday, how likely is it that you will follow-up on the information or referral that you received? (Response options were from 1 = "definitely will not follow up" to 5 = "definitely will follow-up" for each e-therapy service accessed.)

Reasons for help-seeking

To determine the treatment and support needs of family members, we asked a series of open and closed questions. Open questions were related to why they were seeking help online and expectations of the services provided. These were:

  • We are going to ask you a few questions about the reasons you sought help online. Firstly, can you tell me the main reason you recently accessed online help?
  • When you contacted the online service yesterday what help did you expect to receive? For example, if you wanted information, what kind did you want? If you wanted someone to talk to, what did you expect them to do or say?

Family members were also asked to provide their views on how families can be better supported when dealing with problem gambling. Specifically we asked:

  • Finally, we are interested in your views on how we can help the families of people with problem gambling better. What services, support or information would you like to have over the next month?

To determine improvements to e-therapy and possible service options for the future, family members were asked to rate the importance of multiple reasons for seeking help. These items were informed by the literature and consensus amongst the research team. Each of these items was rated from (0) "not at all important" to (3) "extremely important". Items included:

  • Get my family member to reduce their time or money spent gambling.
  • Understand more about problem gambling.
  • Improve my skills in responding to the problem, like assertiveness skills.
  • Get my family member into treatment.
  • Increase my knowledge of support and help options.
  • Get help to better support my family member.
  • Improve the quality of my relationship with my family member.
  • Make my relationship with my family member less stressful.
  • Get help in managing a crisis situation related to the gambling.
  • Improve my skills in managing my emotions or feelings.
  • Have someone to listen to my story and needs.
  • Read stories of people in a similar situation to me.
  • Talk with someone who is in a similar situation to me.

Statistical analysis

Descriptive statistics were used to report on the characteristics of family members, the impact of gambling on family members, types of low- and high-intensity services accessed, types of e-therapy accessed and reasons for help-seeking.

The three open questions: (1) "Can you tell me the main reason you recently accessed help online?", (2) "When you contacted the online service yesterday what help did you expect to receive?", and (3) "What services, support or information would you like to have over the next month?", were analysed in Microsoft Excel using thematic analysis as outlined by Braun and Clarke (2006). This involved (1) familiarisation with data, (2) generating initial codes, (3) searching for phenomena among codes, (4) reviewing phenomena, (5) defining and naming phenomena, and (6) producing the draft results.