Introduction

Prevalence of Gambling Problems

Studies in various countries using a variety of assessment instruments indicate that the past year prevalence of gambling problems ranges from 0.3% (Sweden) to 5.3% (Hong Kong) (Wardle et al. 2011). In North America, estimates of past year problem gambling are very similar, ranging from 0.3 to 3.5% but typically clustering around 1% (Wiebe and Volberg 2007; Alberta Gaming Research Institute 2011; Wardle et al. 2011). In addition, up to 14% of North American adult populations can be considered “at-risk” gamblers (Alberta Gaming Research Institute 2011).

Seeking Help for Gambling Problems

The majority of gamblers do not seek treatment (Cunningham 2005; Petry 2005). In two large U.S. surveys, the Gambling Impact and Behavior Study (GIBS) and the National Epidemiological Survey on Alcohol and Related Conditions (NESARC), 7.1 and 9.9% respectively of pathological gamblers had received professional treatment or attended self-help groups for their gambling problems (Slutske 2006). Approximately 10% of combined problem and pathological gamblers in a California survey reported seeking professional help or going to a Gamblers Anonymous (GA) meeting because of their gambling difficulties (Volberg et al. 2006). Among problem gamblers identified in an Australian epidemiological survey (Slutske et al. 2009), 19.2% had sought gambling-related help, either from a professional or from GA. In Ontario, Canada, 29% of pathological gamblers and 10% of problem gamblers had used treatment services for gambling issues or had attended GA (Suurvali et al. 2008).

Reasons for Seeking Help for Gambling Problems

A number of empirical studies (reviewed by Suurvali et al. 2010) have examined gamblers’ reasons for making a change in their gambling, with or without formal help. As presented in the review, the most commonly cited motivators specifically for seeking help were financial difficulties and relationship issues (including difficulties with spouse or other family members as well as confrontation with or encouragement from other people), followed by negative emotions (such as depression, shame, anxiety, suicidality, and the feeling of having “hit rock bottom”) and then by work or legal difficulties. Help-seeking was also sometimes attributed to physical health problems, conflict with self-image or goals, a desire to regain control, a specific event, or a process of making a conscious decision about one’s gambling.

The Role of Gambling and Demographic Characteristics

The review of motivator studies (Suurvali et al. 2010) revealed two studies focusing on help-seeking which tried to explore factors that might be associated with the choice of motivator. Pulford et al. (2009b) found that compared with gamblers currently accessing a gambling helpline service, gamblers not currently seeking help volunteered significantly more motivators and were significantly more likely to identify most of the specific motivators included in the researchers’ list. However, both groups of gamblers were likely to endorse financial problems as the main reason for seeking help. Evans and Delfabbro (2005) identified no significant differences in motivators between gamblers who had sought professional help and those who had tried primarily self-help strategies. The top three reasons for help-seeking in the combined sample, based on closed-ended questioning, were concerns about health, especially mental health, financial worries and fears of losing one’s home, while open-ended questioning revealed financial issues, mental health and relationships to be the main motivators. The authors also noted that living alone versus living with a partner or family did not produce a significant difference in motivators for seeking help (Evans and Delfabbro 2005).

Overall, though, little research was found on factors that might be differentially associated with gamblers’ expressed reasons for seeking help (Suurvali et al. 2010). No help-seeking motivator studies considered the possible effect of gambling problem severity or gamblers’ own perception of whether or not they had a problem with gambling. In most of these studies, the motivator questions were asked only from those gamblers in the sample who explicitly self-identified themselves as problem gamblers (Tremayne et al. 2001; McMillen et al. 2004; ACNielsen 2007) or who reported having used treatment or self-help to cope with gambling difficulties (Evans and Delfabbro 2005); one can assume that most gamblers seeking assistance with their gaming behaviour would have viewed themselves as having a problem with gambling. No studies were found either that examined the possible role of gender, age, cultural background or socioeconomic measures such as employment status or education in the kinds of motivators for help-seeking identified by gamblers; the authors of the review concluded that more research was needed to address these themes (Suurvali et al. 2010).

The Role of Awareness of Services

Triggers for seeking help have also not been studied in conjunction with awareness of the help available for gambling problems. In over half of the studies asking gamblers about their barriers to seeking help (reviewed by Suurvali et al. 2009), concerns about treatment itself, including lack of knowledge about what treatment was available as well as worries about the quality and efficacy of treatment, were reported as a deterrent to getting help. Problem gamblers and family members of problem gamblers interviewed in the Australian Capital Territory by McMillen et al. (2004) described seeking help when they came upon ads and brochures on treatment services. In Tremayne et al.’s (2001) study, also conducted in the Australian Capital Territory, 79% of problem gamblers receiving assistance learned about treatment services by word of mouth or by asking someone for help, while 11% saw information ads at gambling venues and 10% searched in the telephone directory.

Linking Barriers to and Motivators for Seeking Help

Although many studies have explored the barriers reported by gamblers to seeking help (Suurvali et al. 2009), and all five of the motivator studies specifically examining reasons for seeking help (Suurvali et al. 2010) also addressed barriers to seeking treatment among the same samples,Footnote 1 none have attempted to directly link motivators and barriers. Across studies, the most commonly reported barriers to seeking help were the gamblers’ need to handle their problems by themselves, concerns about stigma, and difficulties admitting the existence or magnitude of the gambling problem. Practical issues associated with attending treatment and concerns about treatment itself were also quite common (Suurvali et al. 2009).

Only in a few of the barriers studies was there some examination of differences between those eventually overcoming barriers and actually seeking treatment and those either relying only on self-help resources or avoiding any kind of help. Cooper (2001, 2004) found that disordered gamblers who had never attended face-to-face treatment or GA were more likely than gamblers who had experienced one or both of these forms of assistance to identify stigma-related barriers. Evans and Delfabbro (2005) observed that disordered gamblers who had relied mostly or completely on self-help methods were significantly more likely than seekers of formal help to fear loss of the social and recreational benefits of gambling as well as to dislike organizations with religious affiliations. In Pulford et al.’s (2009a) study, gamblers currently accessing a gambling helpline service were significantly less likely than gamblers not currently seeking help to subscribe to any of the possible barriers to help-seeking except desire for self-resolution/pride and feeling too overwhelmed by problems. Although both Evans and Delfabbro (2005) and Pulford et al. (2009a, b) examined motivators as well as barriers among their study samples, they did not report on relationships between specific barriers and motivators. The authors of the review on motivator studies (Suurvali et al. 2010) suggested further research on the interactions between barriers and motivators. With the goal of making assistance more available and appealing to at-risk and disordered gamblers before they experience too many serious negative consequences, the design of educational messages and intervention strategies would benefit from knowledge about the factors that permit people with gambling problems to overcome specific obstacles to getting help.

Objectives of the Current Study

Using a representative population survey methodology in order to obtain a sizeable sub-sample of gamblers with possible gambling-related problems, the current study was undertaken to explore how motivators for seeking help might relate to barriers to seeking help, awareness of gambling services and of how to access them, gambling problem severity, self-perception of a gambling problem, history of help-seeking and demographic characteristics among gamblers in Ontario, Canada.

Methods

Survey Sample

A random digit dialing telephone survey study of gambling-related views and behaviours among Ontario adults was conducted between September, 2006 and August, 2007. Stratification by public health region helped guarantee sufficient representation from more sparsely populated parts of the province. Within the randomly selected households, study involvement was offered to the English-speaking person aged 18 or older with the most recent birthday. Interviews were done over the telephone and consisted of both closed and open-ended questioning, with the latter being used to access top-of-mind thoughts about motivators for and barriers to seeking help and awareness of resources and strategies for help-seeking. Open-ended responses were tape-recorded; recordings were destroyed once transcription was completed. All potential study participants provided verbal consent prior to completing the survey. Approval for the study was provided by the Research Ethics Board at the Centre for Addiction and Mental Health.

The study obtained a response rate of 51.7% of eligible participants; this translated into 8,467 interviews. For more information on the survey, please see the full final report (Cunningham et al. 2008).

Measures

Identification of Survey Respondents with Possible Gambling Problems

A sequence of screening questions was used to identify respondents with possible gambling problems, past or present. Respondents who acknowledged having ever spent more than $100 on gambling (the first screener question) next answered the three CLiP (examining Control, Lying and Preoccupation) questions, presented in terms of lifetime gambling history, from the NODS-CLiP (Toce-Gerstein and Volberg 2004; Toce-Gerstein et al. 2009). The CLiP has been shown to have high sensitivity and specificity for NODS constructs, permitting rapid identification of almost all the lifetime pathological gamblers and most of the problem gamblers diagnosed by the full NODS (Toce-Gerstein et al. 2009). More will be said about the full NODS shortly. In the present study, respondents who scored one or more on the lifetime CLiP were also asked the same three questions in the context of the past year. The 730 survey respondents who were identified through this process as possibly having a gambling problem in the past year formed the study sample.

Demographic Information

The interview included questions for ascertaining respondents’ demographic characteristics. Year of birth was used to calculate age and other questions addressed educational background, employment status and marital status. Gender was appraised and coded by the interviewer.

Self-Perception of Gambling Problem

The 730 respondents were asked: “During the past year, have you ever been concerned that you might have a problem with gambling?” Those who replied “yes” or “maybe” were then asked to describe their gambling activities during the past year in terms of six response options, ranging from a low of “not at all a problem” to a high of “a very major problem”. Respondents who said that their gambling had presented at least a “moderate problem” for them, were considered to self-perceive a gambling problem. Those who chose lower impact descriptors for their gambling were considered not to self-perceive a gambling problem.

Severity of Gambling Problem

The seventeen questions of the National Opinion Research Center DSM-IV Screen for Gambling Problems (NODS) (Gerstein et al. 1999), presented in the context of past year gambling experience, were used to classify the 730 respondents (all of whom scored 1 or more on the past year CLiP) as: (1) those who did not qualify on the NODS for any of the past year problem categories (NODS score of 0); (2) those qualifying on the NODS as at-risk gamblers (scores of 1 or 2); (3) those qualifying on the NODS as problem or pathological gamblers (scores of 3 and over). The NODS originated from the need to have a tool based on the diagnostic criteria in the DSM-IV for the identification of problem and pathological gamblers in the general population. It has been tested in the US and Canada and has been shown to have adequate to good reliability and validity (Gerstein et al. 1999; Hodgins 2004; Wickwire et al. 2008; Toce-Gerstein et al. 2009).

Awareness of Services Offering Help for Gambling Problems

The 730 respondents were asked three questions to ascertain awareness of gambling-related assistance services: (1) could they think of any services available to help problem gamblers (if necessary, the respondents were prompted to consider what services might be available in their own community)?; (2) had they ever seen advertising for services to help problem gamblers in their community?; (3) if they wanted to do so, how would they go about searching for gambling help? All three questions were open-ended and the answers were tape-recorded, transcribed and coded by one of the authors. Then, for each question, the responses were dichotomized into those that did and those that did not indicate some awareness of services.

Predicted Reasons for Hesitating to Seek Help for Gambling Problems

Next, the respondents were asked: “If you decided to seek help for gambling problems right now, what type of concerns would you have that might make you hesitate? Can you think of any reasons why you would not want to seek treatment?” Once again, responses to this open-ended question were tape-recorded, transcribed and coded by one of the authors using the following process: From each respondent’s answer, all reasons for hesitation mentioned were noted. Whenever a previously unidentified reason for hesitation was found, the coder returned to the responses that had already been examined to check for the possible presence of this reason. Once all reasons for hesitation had been identified in all respondents’ answers, the reasons were grouped into categories based on common themes; this was done in collaboration with another author. Categories endorsed by 40 or more respondents were included in this study.

Predicted Reasons for Seeking Treatment for Gambling Problems

Respondents were then asked: “Is there any type of problem that might develop in relation to your gambling that would lead you to seek treatment?” Those who replied “yes” were then asked to elaborate on what types of problems these might be. As before, responses to this open-ended question were tape-recorded, transcribed and coded, following the same process as was used for predicted reasons for hesitating to seek treatment. All categories of predicted reasons for seeking treatment are reported in this study.

Previous Efforts to Seek Help

Questions were asked about any experiences respondents had ever had with seeking gambling-specific help from each of the following services: GA, gamblers telephone helpline, inpatient or residential treatment, assessment or outpatient services, family or marital counseling, professionals (physician, psychiatrist, psychologist, social worker) seen at their private office, minister (or priest, rabbi, clergy or spiritual leader), the Internet or printed self-help materials, and other. Respondents were dichotomized into two groups: those who had, at some point in their lives, sought gambling-related help from any of these sources and those who had not.

Analyses

To improve the representativeness of the study results, weights based on age and sex distributions in census data were applied to the current data. All means and proportions reported in this paper are derived from weighted data. However, numbers of respondents are calculated with unweighted data.

A count was done of the number of respondents volunteering each type of problem that might lead to treatment-seeking. Those who said that they could not think of anything that might precipitate help-seeking were also counted. Possible reasons for help-seeking cited by 40 or more respondents were then examined in terms of: three categories of past year gambling problem severity; whether or not the respondents perceived themselves to have had a moderate or greater problem with gambling in the past year; whether or not the respondents had ever sought help for a gambling problem; awareness or not of available helping services; acknowledgement or not of having seen advertisements for services to help problem gamblers; ability or not to suggest at least one way of seeking gambling help if such was desired; most common reasons for possibly hesitating to seek treatment; gender; age groupings; marital status (married or with a common-law partner versus no partner); whether or not the respondents were employed; whether or not the respondents had received at least some post-secondary education. Differences were assessed for significance with Pearson Chi-Square tests.

Results

Characteristics of Study Sample

The study sample (N = 730) was 54.3% male (n = 377), with a mean age of 45.3 (±14.8) years. Approximately two-thirds (67.8%) were employed full-time or part-time, 52.7% had at least some post-secondary education, and 64.7% were married or in a common-law relationship.

Of the sample, six people described their gambling as “not at all a problem” and 43 did not agree to the tape-recording of their answers to the open-ended questions; these people were not asked the questions on reasons for possibly seeking treatment, reasons for hesitating to seek treatment and awareness of helping resources and strategies. Another 155 respondents who did answer these questions had to be dropped from the analysis due to serious static problems on the telephone lines which made it impossible to transcribe some or all of their recorded responses. This left 526 respondents for whom there were usable transcribed answers to all the open-ended questions on motivators, barriers and awareness of resources. The demographic characteristics of the 526 respondents were very similar to those of the original sample of 730 respondents, with the biggest difference occurring in the proportions that were married or had a common-law partner (67.0% of the 526 versus 64.7% of the 730).

Reasons for Possibly Seeking Treatment for Gambling Problems

Table 1 shows the proportions and numbers of participants volunteering possible gambling-related issues that might motivate them to seek treatment. Over two-thirds of the 526 participants could not think of any gambling-related concern that might develop that would prompt them to seek treatment. A few people in this group emphasized that their inability to think of such an occurrence was because they did not have a problem with gambling.

Table 1 Number of respondents who volunteered each reason why they might seek treatment (N = 526)a
Full size table

Among the 154 respondents who were able to think of something that might motivate them to seek treatment, most (n = 112) cited only one motivator. Another 36 respondents cited two motivators and 6 respondents named three motivators. The most frequently predicted trigger for seeking treatment (mentioned by almost a quarter of the sample) had to do with the impact of gambling on finances. A smaller proportion of the sample cited family or relationship issues and an even smaller proportion expressed worry about losing control over their gambling or about the increasing predominance of gambling in their daily lives. A few people mentioned concern about the effect of gambling on their job, their school activities or some unspecified aspect of their lives or the possibility of gambling leading to other problematic behaviours (such as illegal activities, use of alcohol or drugs, lying). Explicit references to negative feelings, health effects and belief systems were very infrequent.

Role of Gambling-Related Characteristics in Reasons for Possibly Seeking Treatment

Two categories of concerns that might trigger gambling-related treatment-seeking were volunteered by 40 or more gamblers in the sample: financial issues and family or relationship issues. In addition, a third category was comprised of the many respondents who said they could not identify any concerns that might lead to treatment-seeking. The proportions of various gambler groups who provided responses fitting into these three categories are shown in Table 2.

Table 2 Reasons for possibly seeking treatment by different gambler, awareness and demographic characteristics (N = 526)
Full size table

Gamblers in the study sample who did not meet NODS-CLiP criteria for problem or pathological gambling in the past year were significantly more likely than gamblers who did meet these criteria to say that they could not think of any possible reasons to seek gambling-related help. Similarly, gamblers who did not perceive themselves as having at least a moderate gambling problem were significantly more likely than those who did perceive themselves as having a gambling problem to be unable to think of something that might prompt gambling-related help-seeking. Conversely, gamblers qualifying on the NODS-CLiP as problem or pathological gamblers or at-risk gamblers, and those self-perceiving a moderate or major gambling problem were significantly more likely than their counterparts with less indication or perception of gambling problems to identify financial difficulties as a possible motivator.

Role of Resource Awareness in Reasons for Possibly Seeking Treatment

Of the 526 study participants, 60.2% reported having seen advertisements for helping services for problem gamblers, 65.6% were able to name one or more such services and 88.8% identified a resource or strategy that they could use if they wanted to look for help for a gambling problem.

Table 2 maps the proportions of gamblers with higher versus lower resource awareness onto the three categories of motivator responses. Those admitting no knowledge of available gambling-related services and those stating they would not know how to proceed in seeking help were significantly more likely to provide no reasons that might motivate them to seek help. Citing at least one service option or strategy for seeking help was, however, associated with a significantly greater probability of volunteering financial motivators. Family/relationship issues were also more likely to be offered by gamblers who knew of available service options. Recollection of having seen advertisements for gambling-related assistance was not significantly related to any of the categories of motivator responses; however, the trends were in the same direction as with the other two awareness measures.

Role of Demographic Characteristics in Reasons for Possibly Seeking Treatment

The associations between demographic measures and the three categories of motivators are also depicted in Table 2. Gamblers with no post-secondary education and those 40 through 49 years old were significantly more likely than gamblers with higher education and those who were younger or older to claim they could not think of anything that might prompt them to seek gambling-related help. Gamblers who were employed full- or part-time were significantly more likely than gamblers who were not employed to identify family or relationship issues as a possible trigger. None of the demographic measures significantly predicted the volunteering of financial issues as a possible motivator; however, age younger or older than 40 through 49, and having at least some post-secondary education approached significance as predictors. No gender differences in identification of motivators were observed.

Relationship Between Reasons for Hesitating to Seek Treatment and Reasons for Possibly Seeking Treatment

Gamblers identifying any of the three main types of reasons for delaying or avoiding seeking treatment (shame, issues with treatment, issues with acknowledging the problem or initiating action) were significantly less likely to report that they could not think of anything that might motivate them to seek gambling-related help (see Table 3). Those who asserted that they saw no reason to hesitate once they had decided to seek treatment were significantly more likely to perceive no reason for actually seeking help. Significant relationships occurred between acknowledgement of all three of the main types of reasons for hesitation and the volunteering of financial issues as a possible motivator, as well as between failure to say there was no reason to hesitate and the volunteering of a financial motivator. Similar but weaker relationships were found with family/relationship issues as the possible motivator.

Table 3 Reasons for possibly seeking treatment by reasons for hesitating to seek treatment (N = 526)
Full size table

Discussion

Reasons for Possible Treatment-Seeking for Gambling Problems

The question on motivators in this study was presented in personal but hypothetical terms: “Is there any type of problem that might develop in relation to your gambling that would lead you to seek treatment?” This type of phrasing was intended to make the question appropriate for all gamblers in the sample, including those who may not have encountered difficulties with their gambling (the sample included many gamblers who met only the past year CLiP criteria and not the past year NODS criteria for possible gambling-related problems), those who had problems but were not ready to admit it, those who had never sought help for gambling problems and those who had sought such help. None of the help-seeking studies in the literature review on motivators for seeking gambling-related help (Suurvali et al. 2010) combined a sample consisting of both help-seekers and non-help-seekers with a question directing respondents to focus on their own situation and on their own possible or actual reasons for help-seeking. With one exception, all of these other studies presented the motivators questions only to gamblers who had already sought help (treatment or self-help). Only Pulford et al. (2009b) also asked a group of gamblers who were not currently seeking help (previous experiences with help were not ascertained) to suggest reasons “someone with a gambling problem” might seek help.

Despite the effort made in this study to elicit possible reasons for help-seeking from as many respondents as possible, a majority said that they could not think of any future gambling-related concern that might lead them to seek treatment. Whether or not these respondents actually had difficulties arising from their gaming practices, they seemed unable or unwilling to contemplate the possibility of seeking help.

Consistent with previous research, among study respondents who did identify a possible trigger for gambling-related help-seeking, financial issues and relationship issues were the most frequently cited. The third most common or most important motivator determined through the literature review (Suurvali et al. 2010) was negative emotion; in the current study only two respondents mentioned this motivator. Though falling well below the criterion of forty mentions for inclusion in the detailed analyses, loss of control was quite prominent as a motivator in this study. All of the motivators volunteered in the current study have also been identified in at least some previous studies of help-seeking motivators (Suurvali et al. 2010).

Overall, other than the “no reason/no problem” response, all of the possible triggers for help-seeking given by the respondents in this study could be classed as negative consequences of gambling. This is in accordance with the results of the literature review (Suurvali et al. 2010), in which gamblers asked about reasons for seeking help primarily invoked harms that had resulted from their gaming.

Gambling Problem Severity

Greater gambling problem severity and self-perception of gambling problems were associated with increased likelihood of volunteering motivators, especially financial ones, for help-seeking. This result makes intuitive sense: gamblers with more severe problems are more likely to have experienced gambling-related consequences that they could not easily ignore and to have given some thought to taking action with respect to their gambling. Some may have considered precisely the question that was posed to them in the study: what kind of gambling-related event or development would it take to propel them to seek help. In addition, in another study based on the same survey results as were used in the current study greater problem severity (as assessed by the NODS-CLiP) was found to be associated with increased likelihood of actually seeking treatment or using self-help resources (Suurvali et al. 2008). This finding has been echoed in other studies (Productivity Commission 1999; Tremayne et al. 2001; Slutske 2006; Slutske et al. 2009). Thus, gamblers with more severe problems not only are more able to identify reasons for seeking help, they are also more likely to actually seek help, possibly because their problems are more likely to have exceeded their tolerance levels (Evans and Delfabbro 2005).

The relationship between the two indicators of gambling problems, the NODS-CLiP, one of many possible instruments for assessing disordered gambling (Shaffer et al. 2004), and self-perception of a gambling problem warrants some further discussion. In a paper focusing in greater depth on the barriers to treatment among the same sample of gamblers as used in the current study (Suurvali et al., in press), the authors noted that over half of the NODS-diagnosed possible or probable pathological gamblers did not perceive themselves as having even a moderate gambling problem. This finding is not unique. For example, among the studies included in the review on motivators for seeking gambling-specific help (Suurvali et al. 2010), Tremayne et al. (2001) found that a quarter of gamblers identified on the South Oaks Gambling Screen (SOGS; Lesieur and Blume 1987) as having severe problems believed they had no problem with gambling. In several other studies, from 20 to 77% of disordered gamblers [determined according to various instruments such as the SOGS and the Canadian Problem Gambling Index (CPGI; Ferris and Wynne 2001)] did not consider themselves to have a problem with gambling (Wynne et al. 1994; Wiebe et al. 2006; Focal Research Consultants 2008; Slutske et al. 2009).

Various reasons may account for the discrepancy between instrument-assessed disordered gambling and self-perceived gambling problems, e.g., limitations of the instruments themselves (Petry 2005; Hodgins and Stinchfield 2008), unwillingness to admit to gambling difficulties, and lack of awareness of the signs of gambling problems and even of what constitutes gambling. With regard to lack of awareness about gambling and its possible consequences, Turner et al. (2005) found, in an Ontario telephone survey of a probability sample of adults, that although almost 90% of respondents were able to identify at least one valid sign of gambling problems, most respondents could not provide more than one sign. In addition, identification of various common games and other activities as gambling ranged from over 97% (for casino table games) to less than 50% (for raffle tickets) and even less than 30% (for long-term investments). With the exception of slot machines, respondents were more likely to identify an activity as gambling if they themselves did not participate in it. Lange (2001) noted that undergraduate university students in Massachusetts seemed not to see themselves as gambling unless they played often, participated in several types of games, or were involved in illegal gambling.

Another possibility is that the trajectory of gambling disorders involves recognition of specific negative consequences at an early stage of problem development (e.g., “I am spending too much money”), which is followed by conscious or semi-conscious efforts to reduce the harm. Individuals at this stage would score positively on problem gambling severity screens without necessarily seeing themselves as “having a problem”. However, failure to reduce this harm despite repeated efforts may then lead to the individuals beginning to re-define themselves as “having a problem”. At this next stage, they would not only score positively on severity screens but would also endorse items asking specifically about having a problem. Evidence in the literature (e.g., Evans and Delfabbro 2005; Pulford et al. 2009b) already suggests that for many disordered gamblers, the harm has to have become very severe in order for them to recognize or admit that they have a problem. If confirmed with longitudinal research, the proposed model of how disordered gamblers arrive at the understanding that they have a gambling problem has implications for the focus of public awareness campaigns in the gambling field. Such campaigns should emphasize the significance of signs of harm as indicators of a problem that needs to be addressed, with the goal of encouraging earlier self-awareness that can then prompt earlier treatment-seeking.

Past Experiences with Help-Seeking

Intuitively, one would expect actual help-seeking experiences to have an impact on the readiness with which gamblers would come up with reasons for possibly seeking help in the future. Although a somewhat higher proportion of those who had never sought help said they predicted no reason for seeking help and a somewhat higher proportion of those who had sought help said they thought financial reasons would prompt them to seek help, the differences were not significant.

It had been necessary to select lifetime help-seeking and not past year help-seeking as the help-seeking experience measure for exploring motivators because of the small number of gamblers (n = 11) who had sought help in the past year. Additional exploratory analyses revealed that only a quarter of the past-year help-seekers versus almost three-quarters of the past-year non-help-seekers (n = 515) said they could not think of any development that might make them seek help; these proportions were reversed for the identification of financial motivators. Although these results are only suggestive and require validation in further research with larger numbers of past-year help-seekers it seems likely that gamblers who have recently sought help for a gambling problem will more readily consider seeking help again (assuming their experiences with help-seeking were not very negative), especially in response to mounting financial difficulties. Help-seeking done longer ago may have less impact as recollection of and familiarity with the process of getting help fades.

Awareness of Services and of How to Proceed with Getting Help

The majority of study respondents were aware of services to help those with gambling problems, had seen advertisements for such services and knew how they might go about looking for help. The proportions found in this study were similar to and sometimes considerably better than the proportions found in other studies exploring awareness of gambling-related media campaigns and of resources to assist those with gambling problems (Najavits et al. 2003 in Indiana; Turner et al. 2005 in Ontario; Volberg et al. 2006 in California; Focal Research Consultants 2008 in Nova Scotia; Ipsos Reid Public Affairs and Gemini Research 2008 in British Columbia; Moodie 2008 in Scotland; Marketquest Research 2009 in Newfoundland and Labrador; Marketquest Research 2010 in New Brunswick).

Awareness figures reflect the reality of what and how many types of services are in the community, how broadly they are publicized, how long these services have existed and been advertised. Differences exist between places and over time. An especially useful comparison for the present results is with the earlier Ontario study (Turner et al. 2005), conducted approximately 6 years before the current study: in that study, 43% said they had seen some kind of public notice about problem gambling and 34% were aware of initiatives to reduce problem gambling. Another important factor in awareness is the salience of the information to the recipient. Turner et al. (2005) interviewed a general population sample; there was no screening based on respondents’ own gambling behaviour. In the current study, the awareness questions were presented only to gamblers and specifically to those meeting criteria for possible gambling-related problems. These people may have been more likely to notice and remember publicity and information dealing with gambling concerns. This suggestion is supported by indications from surveys conducted in British Columbia (Ipsos Reid Public Affairs and Gemini Research 2008), Nova Scotia (Focal Research Consultants 2008) and New Brunswick (Marketquest Research 2010) that awareness of services for those with gambling problems and of advertising for such services is lowest among non-gamblers and often highest among problem gamblers. The exceptions to this pattern, such as the greater recall among at-risk gamblers than among problem gamblers in Nova Scotia (Focal Research Consultants 2008) of social marketing messages on gambling risk recognition and problem avoidance are a reminder of another possibility, namely that some gamblers with problems they are not ready to acknowledge may go out of their way to avoid seeing or remembering anything that might prompt thoughts of possible need for change in their gambling behaviour.

Gamblers who reported knowledge of available services and described how they might go about looking for help were more likely to suggest reasons for seeking help and less likely to provide a “no reason” response. Lack of knowledge about available services has regularly been identified as a barrier to seeking treatment for gambling difficulties (Suurvali et al. 2009). Gamblers with this knowledge have one less obstacle in their path to getting treatment or other kinds of gambling-related assistance. Knowledge of resources and strategies may make the idea of getting help less daunting; it may then be easier to imagine situations that could lead to help-seeking. Among some disordered gamblers, knowledge of resources and strategies may be the product of research on the topic, a possible step towards actually getting help. Additional analyses of the data from this study also revealed that prior experience with helping services was associated with greater knowledge of resources and strategies.

Demographic Characteristics

Employed gamblers and those with better education were more likely to volunteer motivators for help-seeking. Separate analyses showed that gamblers with these characteristics were also more likely to be aware of available helping resources. This result is in agreement with Turner et al. (2005) who found that people with higher education were significantly more likely to be aware of problem gambling-related initiatives in Ontario. As education and employment/unemployment (along with other factors such as income, race, ethnicity) are indicators of advantaged versus disadvantaged social groups (Braveman and Gruskin 2003; Beacom and Newman 2010), with higher education and employment signaling advantage, this finding could suggest that socially advantaged problem gamblers are more comfortable with investigating the services available for their gambling problem and with considering getting help for themselves.

How people interact with health information can be mapped onto a continuum, with active information-seeking at one end and information non-seeking and avoidance at the other end (Beacom and Newman 2010). Research has indicated that non-seeking of health information is highest among groups characterized by low education, low income and old age. Health information has generally become more readily available with the proliferation of health-related websites, but here too, the less educated, lower income, older people are less likely to have computer access, to know how to use the technology properly, and to seek health information on-line (Beacom and Newman 2010).

The disadvantage associated with less education and lack of employment has been shown, in some studies, to extend to attitudes about help-seeking for health issues and to actual use of health-care services. For example, Jagdeo et al. (2009) examined data from two large population surveys, one in the United States and one in Ontario, finding that negative attitudes towards help-seeking for mental health problems were associated with less education. In the U.S., delays in initiating first treatment contact for a mental disorder tended to be shorter for college graduates than for those with less education (Wang et al. 2005). On the other hand, higher use of GP consultation services in Great Britain was associated with unemployment and with lower social class; use of preventative services, however, was lower among those of lower social class (Campbell and Roland 1996). Similarly, Irish men who were unemployed or on sickness or disability leave were more likely than their employed or retired counterparts to seek help for mental health problems; however, men who had only completed secondary school were less likely to seek such help than men who had received higher education (Doherty and Kartalova-O’Doherty 2010). Furthermore, according to a Canadian prospective cohort study with a sample of over 14,000 people, those with less education (and those with lower incomes) were significantly more likely to use primary health care services and equally likely to use preventative care services compared with their more socially advantaged counterparts (Alter et al. 2011). The apparent inconsistencies in results from the various studies may stem from variations in study methodology; they may also reflect true differences among locations and among the kinds of health issues (with corresponding care) being considered.

This raises the question of what role social advantage/disadvantage might play in the use of gambling-specific helping resources. In the current study, additional analyses did not reveal any relationships between lifetime treatment (including self-help) use and education or employment. However, numbers of gamblers with treatment experience were small, and educational level and especially employment at the time of the survey were not necessarily the same as in the past when the treatment was sought. More research is needed to explore this possible relationship.

Studies have also linked measures of social advantage/disadvantage with various health outcomes. For example, Braveman et al. (2010) examined disparities in indicators of health in the U.S. by selected measures of advantage/disadvantage; gradients for educational attainment, with better outcomes associated with more education, were found for life expectancy at age 25, activity limitation, obesity, diabetes and health status. The education-based gradients in health have been found in other countries as well, such as Great Britain and Sweden (Marmot 2006) and Canada (Alter et al. 2011). Disordered gambling has also been found to occur with more frequency among those with less education (Volberg 1994; Wynne 2002; Marshall and Wynne 2003; ACNielsen 2007; Rush et al. 2007; Wardle et al. 2007; Lemaire et al. 2008), although not in all studies (Wiebe et al. 2006; Focal Research Consultants 2008; Ipsos Reid Public Affairs and Gemini Research 2008; Wardle et al. 2011). Similarly, unemployment tends to be associated with a higher incidence of disordered gambling in many studies (Cox et al. 2000; Smith and Wynne 2002; Wynne 2002; Delfabbro and LeCouteur 2003; ACNielsen 2007; Rush et al. 2007; Focal Research Consultants 2008; Ipsos Reid Public Affairs and Gemini Research 2008; Lemaire et al. 2008; Wardle et al. 2011).

Braveman et al. (2010) point out that there is considerable evidence to support the argument that education and income affect health through their impact on access to a wide variety of opportunities and resources, only one of which is medical or health care. Marmot (2006) argues for the major role in health of position in the social hierarchy, explaining that within a given society, position in the hierarchy is a determinant of how much control (over one’s life and work) and social participation are available to a person. Low levels of social participation and control have been linked to increased risk of various illnesses. At least partly through its impact on literacy, education functions as a mediator between social position and health; it is also a point for possible intervention (Marmot 2006).

Literacy, including health literacy, is important in how people acquire and use health information (Beacom and Newman 2010). The Institute of Medicine in the U.S. estimates that more than half of the adult population may be health illiterate, and thus unable to find and make proper use of basic health information. Despite the increased availability of on-line health information, the material requires, on average, grade 10 reading skills, and over half of it demands post-secondary level reading skills (Beacom and Newman 2010). There are additional difficulties for newcomers with different language backgrounds, and even the nuances of dialect and cultural meaning can greatly affect the accessibility and meaningfulness of health communications. Beacom and Newman (2010) offer examples of several innovative types of efforts to tackle the health literacy problem, including computer support systems among peers; interactive computer programmes that customize information based on the user’s background; trained lay community health workers who disseminate health information among peers; transmission of health information using formats (such as animation) from the entertainment field; insertion of health education messages into entertainment programming (such as TV shows with a fictional story); they note, though, that more research is required to evaluate the effectiveness of these strategies. Some of these strategies could be promising also for increasing awareness of gambling-related difficulties and of resources that can provide assistance; here too are ideas for research and development.

Barriers to Seeking Treatment

In their study focusing on barriers to seeking help among the same gamblers as comprised the current study sample, the authors (Suurvali et al., in press) proposed that the ability and willingness to identify factors that might elicit reluctance to seek help could indicate acknowledgement of a need to take action about one’s gambling and thought given to potential difficulties associated with this step. A similar relationship has been proposed here with regard to possible motivators. Intuitively it makes sense that disordered gamblers who have invested effort in considering their gambling problem and the need for action would be more likely than those who have avoided or been unaware of these issues to have pondered the specifics of help-seeking, including both the obstacles and that which might override the obstacles. Support for the idea that volunteering barriers and volunteering motivators are linked can be drawn from the finding that gamblers who identified barriers to seeking help were also more likely to volunteer a possible motivator.

In the literature review on motivators, Suurvali et al. (2010) pointed out that research is needed to examine the connection between specific barriers to seeking help and what prompts gamblers to overcome these barriers. For example, are the triggers to actual help-seeking the same for gamblers inhibited by shame as for those worried about how they could find the time to attend treatment? Although the current study did not explicitly ask respondents what would lead to help-seeking despite a specific barrier they had identified, the analyses described here did somewhat address the link between types of barriers and types of motivators, at least at a hypothetical level, in the thinking of gamblers. Respondents were first asked what might make them hesitate even though they had decided to seek gambling-specific help, and then what kind of gambling-related problem they could imagine developing that might lead them to seek treatment. Given the proximity and sequencing of the two questions, respondents may have been influenced in their choice of motivators by what they had recognized as likely barriers.

What emerged, though, from the analyses, was very little difference in the types of motivators expected to push the gamblers past their reasons for hesitation. Whether respondents expected to hesitate because of shame, issues associated with the treatment itself, or difficulties acknowledging the gambling problem/initiating action, roughly a third thought that financial hardships would spur them into action while approximately one out of seven felt that family or relationship concerns might provide the incentive. It is possible that the restriction of comparative analyses to groupings endorsed by at least 40 respondents prevented the discovery of some emergent patterns in the types of motivators and barriers volunteered. However, when the motivators endorsed by fewer than 40 respondents were examined in relation to the corresponding barrier types, no suggestion of links between specific motivators and specific barriers was found.

The current study assessed gamblers’ hypothetical barriers and hypothetical motivators. Although there was a rationale for selecting this type of wording (as explained earlier in the Discussion), it may be fruitful in future research to ask gamblers who have actually sought help quite recently (so that the experience is reasonably fresh in their memories), what obstacles caused them to hesitate and what eventually prompted overcoming these particular obstacles. As population surveys tend to produce very small numbers of gamblers who have recently sought help, it may be best to select a sample of gamblers who have just entered treatment. Open-ended questions would continue to be desirable, encouraging the respondent to provide details on how the barriers were confronted. Coding of the motivator responses should then focus, not only on increasingly insistent gambling consequences but also on the cognitive processes and other strategies used in getting from hesitation to action. The insights resulting from these types of studies, into what helps problem gamblers get past indecision and procrastination when it comes to seeking gambling-related assistance, will be useful in the development and delivery of more accessible and appealing services and resources for people with gambling problems.

Limitations of the Study

Issues with Response Rates and Loss of Respondents

The low response rate (just over 50%) to the overall population survey is a concern. Response rates achieved in telephone surveys generally have been declining, with much of the decline being attributed to less willingness to take part in surveys and the proliferation of technological developments such as Call Display and Call Blocking that facilitate the avoidance of undesired calls (Curtin et al. 2005; Keeter et al. 2006; Volberg et al. 2006). In the current survey, up to twelve call-backs were made at different times of the day, including in the evenings and on weekends, in order to try to increase the chances of making contact with someone associated with each eligible randomly selected telephone number. To maximize survey completion once contact was made, well-trained interviewers were employed to conduct the recruitment and interview process.

Another possible factor in lower response rates in random digit dialing telephone surveys is an increase in telephone numbers associated with non-voice applications (e.g., computers, fax machines) and in unassigned telephone numbers (Curtin et al. 2005). Response rate is broadly defined as the number of completed interviews with reporting units divided by the number of eligible reporting units (American Association for Public Opinion Research 2011). If a telephone number is never answered by a person or answering machine/service, it can be very difficult to determine whether or not it is associated with an eligible reporting unit. There are several alternative ways of treating telephone numbers of unknown eligibility and the choice affects the resulting response rate (American Association for Public Opinion Research 2011). The approach used by the researchers in this study incorporated an estimate of the proportion of telephone numbers of unknown eligibility that are actually eligible.

Although the response rate in the current survey was low, it was comparable with the 61% and 53% response rates obtained, respectively in the 2006 and 2007 editions of the annual Ontario addictions and mental health surveys conducted by the Centre for Addiction and Mental Health (Ialomiteanu et al. 2011). It was also in the same range as or better than the response rates reported in other population survey-based gambling treatment motivators and barriers studies done in approximately the same time period (47.2% for Volberg et al. (2006) in California; 43% for Doiron (2006) in Prince Edward Island; 15% for ACNielsen (2007) in New South Wales). It should be noted, however, that because of the many possible variations in how study samples are designed and obtained and how response rates are calculated, caution is required in comparing response rates across studies (Langer 2003).

Although studies have suggested that low response rates in telephone surveys do not necessarily mean poorer quality results than would be obtained with a higher response rate (Langer 2003; Keeter et al. 2006), insufficient research has been done to indicate that this might be true irrespective of the content of the survey (Keeter et al. 2006). The biggest concern with low response rates is nonresponse bias. With the current study, this could mean, for example, that disordered gamblers with a particular demographic profile or gambling history might have been disproportionately represented among people who never answered their telephones, or who only had a mobile telephone and thus were never called. This would limit the extent to which it is possible to say that the results of this study are representative of all disordered gamblers in Ontario. Although weighting was done to increase the representativeness of the data, the weights were based only on age and sex distribution.

In addition to the overall low response rate for the survey, the current study suffered from some loss of respondents due to objections to the taping of open-ended responses and static problems on the telephone lines. Here too, certain gambler or demographic subgroups may have been more sensitive than others about possible issues associated with tape-recording and even reliance on cordless telephones (which are more prone to static than telephones with cords) could have varied among such groups. Despite these possible sources of skew, however, the agreement between results from this study and those from other similar studies as well as among results from various analyses within this study supports the validity of the observations, at least for some segment of disordered gamblers and suggests intriguing relationships between gambler characteristics and motivators for help-seeking worthy of further research.

Issues with Responses to Open-Ended Questions

Considerable reliance was placed in this study on open-ended questioning. Although responses elicited by open-ended questions can provide important insights into gamblers’ own thoughts about their gambling, they do present challenges for data analyses. For example, it is desirable as a reliability check to have two coders for open-ended material but in this study, resource limitations permitted only one coder at the initial stage of analyzing verbatim responses. This coder was one of the authors of the study, a researcher familiar with previous work on motivators for and barriers to help-seeking as well as on resource awareness. Although this researcher did not try to fit respondents’ answers into any pre-determined framework of possible motivators and barriers, it is inevitable that his research knowledge would have influenced his interpretation of what the respondents said. It is possible, therefore, that despite best efforts, the coding may sometimes have distorted or failed to capture the meaning of some open-ended answers. More desirable would have been to have two independent coders, at least one of them unfamiliar with the research area. Reliability received a boost at the categorization-of-responses stage as consensus was required between two researchers for this task.

The telephone survey also provided limited opportunity for probing responses. It was quite lengthy and not conducive to in-depth exploration of any one topic, even when the questions asked were open-ended. A more relaxed interview environment, encouraging more elaboration of brief answers and clarification of possibly ambiguous ones might have added to the confidence one could have in the coding. It might also have permitted respondents to think of other possible motivators, barriers and help-seeking strategies, in addition to the ones they mentioned, thereby adding to the richness of the data obtained.

Nonetheless, the majority of the respondents’ open-ended answers were clear. Even if some possibly unique motivators were missed because of limitations associated with survey administration and coding, the fact that relatively large numbers of respondents identified motivators that fit into two categories (finances and relationships), categories that have emerged as important in other studies, including ones conducted with different questioning approaches, suggests that this study has presented a valid picture of the kinds of reasons for which many disordered gamblers may choose to seek help. However, the “no reason/no problem” category of motivator response warrants more research, in particular to learn more about what this kind of apparent rejection of help-seeking means: Does it hide possible triggers to help-seeking which might be revealed in a more leisurely, conversation-style interview? Does it reflect a strong belief in making gambling-related changes alone, without formal help? Are there distinct identifiable sub-groups within this category, for example, gamblers who are convinced they do not have a problem, gamblers who know they have a problem but do not want external help, gamblers who have not yet given the question any thought and are not prepared to do so at this time?

A similar summary can be made of the validity of the results on barriers to seeking help, and a similar argument can be made for further research on the “no reason to hesitate” response (see Suurvali et al., in press for more discussion on this topic). The awareness questions present less of a concern in this study; there is considerably less room for individual interpretation or misunderstanding when responses are assessed simply for the presence or absence of the mention of resources, strategies or advertisements.

Issues with Small Numbers

As is often the case with gambling studies, including population surveys, the numbers of gamblers meeting criteria for disordered gambling were quite low; this limited the types of analyses that could be conducted with the data. Despite this shortcoming and the difficulties presented by response rate and data loss issues, it was possible to identify a fairly large group of Ontario gamblers who may have had some difficulties with their gambling in the past year and to explore the reasons they could envision for seeking treatment.

Summary and Conclusions

The results of this population survey study, conducted in Ontario, Canada, supported findings reported in the literature, that gamblers were most likely to cite financial and relationship concerns as reasons to seek help for their gambling problems. A high proportion of study respondents, however, were unable to offer a reason that might prompt help-seeking. Gambling problem severity, knowledge of treatment services and strategies, identification of barriers to treatment and indicators of social advantage were associated with the likelihood of volunteering motivators for help-seeking. The authors suggested that gamblers who can offer possible reasons for seeking help are more likely to be people who know and acknowledge that they have a gambling problem and who have considered what to do about it. More research is needed to explore how gamblers come to understand that they have a gambling problem and how they identify and overcome specific obstacles to getting help. Knowledge about warning signs of impending gambling problems and about resources for assistance are likely to be important, and socially disadvantaged (e.g., less educated, unemployed) gamblers may have greater difficulty acquiring this knowledge. As gambling problems tend to be more prevalent among socially disadvantaged groups, it is important to focus on these groups in order to better understand how social advantage/disadvantage impacts the gambling trajectory so that more effective educational messages and interventions can be designed to reach these gamblers in a timely manner.