The purpose of this paper is to provide an overview of some trends among problem gamblers who sought counselling from the specialist BreakEven/Gambler's Help problem gambling counselling services in Victoria, Australia, between 1 July 1995 and 30 June 2000 (the latest date for which data were available at the time of writing).

As part of the standard operating procedures of the BreakEven/ Gambler's Help problem gambling counselling services, data about clients are collected at registration and assessment and at each client contact. Counsellors record this information on forms during (or immediately after) contact with clients. These forms constitute the Problem Gambling Services minimum data set (MDS). The MDS was established and implemented in 1995, and revised in 1998, to provide information for a range of purposes, including individual case management, service planning, and analysis of gambling impacts. The Problem Gambling Research Program at the University of Melbourne provided annual reports between 1996 and 2001 based on analysis of the MDS, which were published by the Victorian Department of Community Services (Jackson, Thomason, Ryan, & Smith, 1996; Jackson et al., 1997; 1999a; 1999b; Jackson, Thomas, Thomason, Holt, & McCormack, 2000c; Jackson, Thomas, Ross, & Kearney, 2001).¹

In a pattern highly consistent over the entire 1995–2000 period, approximately one quarter of clients presenting in 1999–2000 (23.6%) were born overseas, representing a wide range of countries. This is broadly consistent with Australian Bureau of Statistics (ABS) 1996 Population Census data, which found that 27.5% of Victorians were not Australian born. However, the rates of problem gambling in some ethnic groups studied in the Victorian Casino and Gaming Authority-funded study of the impact of gambling on specific ethnic groups (Thomas & Yamine, 2000) were found to be much higher than for other groups. One might expect much higher rates of presentation to BreakEven/Gambler's Help within these groups than those found in the MDS data. As shown in Table 5, over half of Gambler's Help clients born overseas (56.6%) were born in Europe.

Table 6 reveals that almost three quarters (73.0%) of problem gamblers presenting for help in 1999–2000 were part of the labour force; they were either employed or looking for work. As in Australian society more generally, the labour force participation rate within the study sample is higher among males than females. However, the unemployment rate of 16.6% among Gambler's Help clients was more than double that for all Victorians aged 15 and over (7.3% (ABS, September 1999), x² = 387.0, df = 1, p < .001).

Table 7 shows that the labour force participation rate for problem gamblers increased between 1996 and 2000, and at 73.0% for the 1999–2000 year was considerably higher than 62.9%, the census figure for all Victorians aged 15 and over as at August 1999 (ABS, 1999). This suggests that problem gamblers have higher rates of both employment and unemployment than the overall population of Victoria. This may be a reflection of the fact that people with gambling problems are predominantly of workforce age (15–64 years).

As can be seen from Table 8, employed problem gamblers who presented for counselling in 1999–2000 reported holding a wide range of jobs. When compared to all employed people aged 15 and over in Victoria, the occupational categories of problem gamblers were very similar. The exceptions were that problem gamblers were underrepresented among professionals (9.6% vs. 19.0% of all Victorians) and overrepresented among labourers and related workers (15.9% vs. 9.7%) (ABS, 1999).

The self-reported individual income of problem gamblers is not confined to a particular income level and varies considerably from less than AUS $120 to over AUS $1,500 per week; however, over half (53.5%) of the clients reported weekly incomes of less than AUS $400. This is explained by the fact that 34.5% of the problem gamblers were in receipt of pensions and benefits. A further 23.3% of problem gamblers reported weekly incomes of between AUS $400 and AUS $599. Lower income figures among the female clients were not unexpected: they reflect the situation in the wider community. For example, in May 1998, the ABS reported that the average weekly earnings for male employees were AUS $730 versus AUS $484 for females.

Further details of income reported by problem gamblers are presented in Table 9. It should be noted, however, that income figures do not account for variations such as whether net or gross income was reported or whether mandatory deductions (e.g., Child Support payments) are counted in gross amounts, and so on. The income distribution of problem gamblers is quite similar to that of all Victorians aged 15 and over from the 1996 Population Census figures, as shown in Table 9.

Clients presenting to BreakEven/Gambler's Help are asked to provide information about hours and dollars spent gambling at their most recent gambling session by gambling type and days spent gambling during a typical month by gambling type. While Table 10 shows that problem gamblers report involvement in a wide range of types of gambling, by far the most common is EGMs, which are used by more than two thirds (71.1%) of all problem gambling clients (86.4% of female clients and 58.3% of male clients). EGM use was the most common form of gambling reported by BreakEven/Gambler's Help clients during the entire five-year period. The next most commonly reported type of gambling was offcourse betting at the TAB (11.9%). No other gambling type exceeded 4%. Only a very small number of people reported involvement in forms of Internet gambling, but because it is quite new, there may be a lag period between the introduction of this type of gambling and clients presenting themselves for problem gambling counselling.

These findings concur with previous research findings that problem gambling is typically associated with particular types of gambling, such as EGMs, racing, video-based games, and casino gaming (Blaszczynski, Walker, Sagris, & Dickerson, 1997). The availability of EGMs, in particular, has been shown to have a strong association with problem gambling (Dickerson et al., 1997), and gamblers who regularly prefer racing and gaming machines have been found to be more likely to score in the at-risk categories in the South Oaks Gambling Screen (SOGS) (Lesieur & Blume, 1987). Although surveys indicate that between 30% and 40% of adult Australians play the lottery as a weekly habit, few, if any, manifest signs of associated problems (Blaszczynski et al., 1997).

These findings also lend support to research indicating that females are more likely to gamble on activities based on “chance,” such as EGMs and bingo, and males are more likely to gamble on forms of gambling based on “skill,” such as TAB and racing (Walker, 1992).

Despite the majority of problem gamblers reporting using EGMs on their most recent day of gambling, a number of issues should be taken into consideration in interpreting the data.

Staff in EGM gaming venues are now more aware of BreakEven/ Gambler's Help services as a result of the gaming industry's access to training programs for venue staff, the community awareness and edu-cation campaigns conducted throughout the state by BreakEven/ Gambler's Help community education workers, and the introduction of Responsible Gaming policies in many venues. Thus venue staff are more likely to be in a position both to recognise problem gamblers and to refer them to a BreakEven/Gambler's Help counselling service. Evaluation of a statewide mass media education campaign supplemented with locally distributed pamphlets also showed that venue patrons were aware of a wide range of messages encouraging responsible play and help-seeking. Of 71 venue patrons interviewed, unprompted recall of problem gambling educational material was 67%, while 79% recalled the slogan “If you have a gambling problem in your life call Gambler's Helpline toll free” (Jackson, Thomas, Thomason, & Ho, 2000b). This study also showed a direct correlation between placement of educational materials and increased calls to Gambler's Helpline with a flow-on effect of referrals to BreakEven/Gambler's Help (Jackson et al., 2000b, pp. 37–40).

Although the amount of money and time spent gambling varies between different types of gambling, as Table 11 indicates, problem gamblers report spending considerable amounts of time and money on most forms of gambling.

A problem gambler's involvement in gambling can vary over time, and as Table 12 suggests, BreakEven/Gambler's Help attracts problem gamblers with wide-ranging patterns of gambling. While one quarter (26.5%) presented to Gambler's Help when their gambling was increasing, 28.7% of problem gamblers had already decreased their gambling. In some cases there was a total reduction to the point where the clients were not gambling at all when they sought assistance.

The patterns of gambling behaviour in Chart 1 show a slight trend for people's behaviour to be less stable at the point of referral, and more people tend to describe their pattern as binge gambling or one of increasing gambling activity. As Table 13 shows, for the 1999–2000 cohort, this increase in gambling or binge gambling is likely to be more characteristic of EGM players than those engaged in offcourse betting.

The extent of clients’ gambling-related debts is reported in Table 14 and shows that males report being more in debt as a result of their gambling than females. This is consistent with the data presented in Table 15 that males are more likely to borrow money to fund their gambling. The mean debt decreased from AUS $14,140 in 1997–1998 to AUS $8,295 in 1998–1999 but rebounded to AUS $11,763 in 1999–2000.

Although, as Blaszczynski et al. (1997) have pointed out, it is difficult to treat self-reported expenditure (loss) information with certainty, we suggest that information on debt, while probably underreported, is worth examining in more detail to determine finer patterns. While over a third of BreakEven/ Gambler's Help clients in 1998–1999, for example, had zero debt, there was a small group of people who had considerable debt. Simple addition of the reported debt for 1998–1999 clients resulted in a total debt of AUS $15,342,362 among the client cohort. However, looking at those with large debt, we find that only 1.3% of problem gamblers accounted for 29% of the total debt (26 people owed AUS $4,486,000).

Further disaggregation of this small, high-debt group, compared to all 1998–1999 clients, showed that they were more likely to be male (62%), older (mean age of 41 years), born overseas, employed full time as a manager or clerical worker, with either a very high or a very low individual income. They had higher proportions of most presenting problems, particularly financial and employment problems, and were four times more likely to have the maximum 10 maladaptive behaviours or DSM-IV items. Thus, while it is probably a good sign that some people are attending counselling with zero or low debt, we cannot lose sight of the fact that the problem gambling population includes people who have what can only be described as massive financial problems.

In a specialised problem gambling counselling service it is not surprising that clients presenting on their own behalf are overwhelmingly concerned about their gambling behaviour, as Chart 2 shows. Other presenting problems as reported by about half or more of these clients were financial, intrapersonal, and relationship or interpersonal issues. While some problems were reported by both sexes, substantial differences were revealed in respect to some problem areas.

Employment and other work-related issues and legal issues were more frequently indicated among the presenting problems reported by men, whereas women were more likely to report relationship and family problems and physical symptoms associated with their gambling.

Comparing the presenting problems reported by problem gamblers over the period from 1995 to 2000 reveals some fairly major changes. We can see quite distinct differences between the periods 1995–1997 and 1997–2000. In the 1997–2000 period all presenting problems declined, with the exception of gambling behaviour. There was a marked decline in leisure use issues and physical symptoms, for example.

Generally, the first two years of the BreakEven/Gambler's Help counselling service may be characterised by a high level of help-seeking for “symptomatic” effects of gambling. This declines steadily from 1997 as clients put a somewhat greater emphasis on gambling behaviour as the reason to seek help rather than for problems which are consequences of the gambling behaviour. This is not to suggest a simplistic notion of causality, however, with all problems deemed to be the result of gambling. It is clear that for many people in counselling, the gambling itself is symptomatic of other problems, such as loneliness, poor familial and interpersonal relations, and perceived lack of alternative leisure pursuits.

The number of presenting problems has also decreased over the years. This suggests both that individuals are seeking help from the program earlier in their “career” as problem gamblers and that the “backlog” of people with longer-standing problems, who had not previously sought help, was dealt with primarily in the first two years of the service's operation. Almost one quarter (24.2%) of problem gamblers in 1999–2000 presented with only one, or none, of the problems listed in this chart compared to 2.4% in 1995–1996. Conversely, the proportion of those presenting with six or more problems was down to 18.5% in 1999–2000 from a high of 49.4% in 1995–1996.

The client assessment form completed by the counsellors at BreakEven/Gambler's Help measures the extent to which gambling has become problematic and uses the DSM-IV criteria as part of the measurement of the type, frequency, and intensity of gambling behaviours and their adverse effects.

The proportion of gamblers reporting five or more maladaptive behaviours is reasonably stable at about three quarters for the period from 1996–1997 to 1999–2000 (see Table 15). This contrasts with the 1995–1996 figure of 81.7%, although the latter figure may be elevated due to the same backlog effect noted earlier. As with the number of presenting problems, there has been a decrease in the number of maladaptive behaviours that people present with. In 1995–1996, 18.2% of people presented with fewer than five of these behaviours compared with 25.7% in both 1998–1999 and 1999–2000.³

If we take the numbers of maladaptive behaviours to be a crude measure of acuity, then it would seem that clients’ gambling problems are decreasing in acuity. However, we caution that this may not be a sound index of acuity. The quality and scope of a behaviour rather than whether it is shared with other behaviours might be a better case acuity index. These measures are not available within the current data collection and would need to be the subject of a separate investigation.

As Table 16 demonstrates, the extent to which maladaptive behaviours manifest themselves also varies according to the type of gambling. Comparing people who gamble using EGMs with those who gamble at the TAB, the TAB clients were more likely than the EGM players to report being preoccupied with gambling; chasing losses; relying on others for money; and having jeopardised relationships, jobs, and education. EGM users were more likely to report gambling as a form of escape. TAB clients were almost twice as likely to report committing illegal acts to finance their gambling.

Further disaggregation of the two variables in the MDS relating to illegal actions in relation to gambling shows that males are twice as likely as females to commit such illegal acts. (The variables are illegal action reported as a current source of gambling funds and the DSM-IV criterion of having committed illegal acts.) Problem gamblers who are younger, unemployed, never married, living in a shared household, reporting debt in excess of AUS $10,000, presenting with legal problems, and reporting 10 maladaptive behaviours are over-represented in this area of illegal acts related to gambling.

This paper sought to describe some trends among people seeking help with gambling-related problems and, specifically, trends among clients of BreakEven/Gambler's Help, the specialised statewide counselling service established as part of the PGSS in Victoria, Australia. A number of areas covered by this overview have been analysed in some detail to date. These include

• gender differences in help-seeking and problem presentation (Crisp et al., 2000)
• debt (McCormack & Jackson, 2000)
• the particular issues faced by partners of problem gamblers (Crisp et al., 2001b)
• the relationship between number of sessions of counselling and problem resolution (Crisp et al., 2001a)
• the characteristics of older people's gambling and help-seeking (McCormack, Jackson, & Thomas, 2003)

There are still numerous questions posed by this review, however. Just some of these are as follows:

• Why is the gender profile for presentations to BreakEven/ Gambler's Help different from that recorded by other problem gambling counselling services? Although we know that women are more likely to seek help with problems, this does not, in itself, seem to be a good enough explanation. We believe that the reason may lie in the location of these services as a specialist counselling program in generalist community health agencies rather than being stand-alone problem gambling services. This location philosophy addresses both access and stigma issues.

• Why are the numbers seeking assistance now over two and a half times the per annum referral rate of the first year of operation of the service? There is obviously a growing familiarity with the existence of both the face-to-face counselling service and the Gambler's Helpline telephone counselling service, which we might expect would lead to more referral, particularly because advertising of the service is increasingly a feature of public education campaigns and venue-based social responsibility initiatives. We have explored in a fairly minimal fashion the relationship between education and help-seeking, but we need stronger evidence of this relationship through detailed evaluations of campaign outcomes. A high level of awareness of actual or potential gambling-related harms may also contribute to a readiness to seek help, to the extent that people may feel that they are less personally responsible (morally and behaviourally) for experiencing problems with their gambling, particularly on EGMs. The Seventh Survey of Community Gambling Patterns and Perceptions (Roy Morgan Research, 2000), for example, found that 84% of Victorians thought that gambling-related problems had worsened, while 83% thought that gambling was now a serious problem.
  Increased numbers of referrals do not appear to be linked to higher levels of participation in gambling in the general community or to higher levels of participation in forms of gambling associated with greater risk of problem play, such as EGMs.
• Is the fact that people are referring with fewer problems an indication that community education and the operationalising of beliefs about harmfulness, as noted above, are a sustainable trend towards early intervention? If this is the case, it would be helpful to know what the policy and program mix was that has achieved this result, what the interplay between government regulation based on harm minimisation and industry social responsibility measures is, and whether the effect could be enhanced with higher levels of either or both.
  We do not know whether fewer problems actually translates into less severity, although, naturally, we would hope for that effect. This still needs to be determined, as does the question of how long people have experienced their gambling problem before seeking help. We are still working on these two important questions:
• How do programs such as Gambler's Help attract more problem gamblers under 25 into counselling?
  How do programs such as Gambler's Help attract non-English-speaking community members into counselling or into more relevant forms of intervention, given that we know that the client participation rate in counselling is almost identical to that of the general adult community profile, whereas we know that problem gambling rates in some of these communities are at least five to seven times those of the English-speaking population?

As suggested, these are just some of the questions that this overview has thrown up, and they are beyond the scope of this paper to answer. There are yet other areas that intersect with research being carried out in other jurisdictions which also need further exploration in the population reported here. Such areas include the DSM-IV profiles of male and female gamblers and also EGM gamblers compared with track (oncourse and offcourse) bettors. What is obvious, however, is the usefulness of “available data,” in the form of an MDS, in gaining a broad picture of help-seeking patterns and in helping to shape a more detailed research agenda in this area of gambling studies.