Gambling participation and rates of problem gambling change over time in response to a variety of factors including gambling availability, demographic changes and adaptation at individual and societal levels. These relationship are complex and only partially understood. The major aim of the present study was to provide general population estimates of gambling participation and problem gambling for Sweden and compare these estimates with estimates from a previous national study. The study was also designed to identify risk factors for problem gambling including change in these factors over time. Data are from the first phase of the Swedish Longitudinal Gambling Study (Swelogs) in which a representative sample of 8,165 people was assessed using validated problem gambling and other measures to facilitate comparison with findings from the 1997/1998 Swedish Gambling Study (Swegs). Overall, it was found that gambling participation reduced markedly, although in some population sectors increases were evident for some forms including poker and electronic gaming machines. Lifetime prevalence of probable pathological gambling increased; however, past 12 months probable pathological and problem gambling prevalence did not. Males, younger adults and people born outside Sweden were at high risk in both studies. Significant prevalence increases were evident for people aged 18–24 and those with low levels of education. The results indicate that relationships between gambling exposure, participation and problems are dynamic with shifting implications for public health and social policy.