Low income and vulnerability for gambling problems.

Abstract

In their editorial that appeared in the June 2004 issue of Addiction, Rehm et al. (2004) cite poverty as an important environmental modifier of the relationship between substance use and harm (the topic of their editorial). For example they cite Harrison & Gardiner (1999) who state that men in the lower socio-economic status (SES) category in their study had a much increased risk of alcohol-related mortality compared to those in the highest SES category, despite the fact that volume of drinking was greater for those in higher SES categories. Readers may like to know that the same effect appears to be evident in the case of gambling and gambling problems according to a secondary analysis of the data from the first British Gambling Prevalence Survey (Orford et al. 2003). The rate of participation in any gambling in the last 12 months and the number of different gambling activities engaged in showed little variation with household income, although both were lowest in the group on very small incomes (less than £5200 per annum). Nor did average total stake on activities such as the national and other lotteries, football pools and bingo or average sum of losses on activities such as playing fruit machines, betting on horse or dog races or playing casino games, vary very much by income (questioning people about expenditure on gambling is fraught with difficulties, and preliminary work suggested that people thought in terms of stakes for certain activities and in terms of losses for others). When stakes and losses were calculated as percentages of income the picture was much clearer, with the highest averages in the lowest of three income categories (less than £15 600 a year in 1999–2000) and lowest in the highest income group (£31 200 or more). That may go a long way towards explaining one of the main findings of the survey, that household income was one of a small number of variables significantly associated with problem gambling in a logistic regression analysis. Controlling for other socio-demographic variables, those in the lowest of the three income categories were nearly three times as likely as the average person to score above the threshold on a problem gambling screening measure. Income appears to be an important modifier of the relationship between extent of engagement in gambling and gambling-related problems.