Dental caries in young children remains a significant public health problem in the United Kingdom. Disease experience in 5-year-olds has remained largely unchanged over the last 20 years and large inequalities are evident between affluent and deprived areas (Pitts et al. 2005). For many years there have been concerns that the majority of disease in the population is untreated (Curzon and Pollard 1997). In 2003/4 only 12 percent of caries in 5-year-olds in England and Wales was treated by restoration (Pitts et al. 2005). What is less widely discussed is that this headline statistic masks large variation in the restorative index at Primary Care Trust level; in 2003/4 the restorative index ranged from 4 percent to 43 percent (Pitts et al. 2005). This large variation was also evident in NHS dental activity data collected by the Dental Practice Board (Dental Practice Board 2005) and cannot be fully explained by variations in disease and service supply at this crude geographical level. Therefore from a public health perspective we have two problems: • large inequalities in dental disease, and • large variation in the amount of restorative care provided to children. To tackle the first problem, we know that fluoridebased interventions are effective in preventing caries (Marhino et al 2002 (a), Marhino et al 2002 (b), Marhino et al 2003). However, primary dental care-based interventions alone won’t reduce whole population disease levels or tackle inequalities in caries levels. Indeed they are more likely to widen inequalities, as they can only reach children who attend the dentist on a regular basis and we know that children who attend sporadically harbour a disproportionately large share of population disease (Tickle et al. 1999, Tickle et al. 2000). Therefore to reduce population disease levels and tackle inequalities, resources for prevention should be focused on effective, fluoride-based, population-level interventions rather than those delivered in primary care. Water fluoridation, for example, reaches attenders and non-attenders alike reducing the disease burden (McDonagh et al. 2000) on primary care services making management of young children easier for General Dental Practitioners (GDPs) (Threlfall et al. in press). We know far less about the second problem; which is why is there such a wide variation in the amount of restorative care provided and how to address this issue. In the UK the greater part of dental care for children is provided by GDPs working in the NHS. Although factors such as access to, and utilisation of dental services are important, it is crucial to understand how GDPs approach the care of young children. A retrospective cohort study reported the outcomes of care delivered by 50 GDPs in the North West of England. (Tickle et al. 2002). When the care provided by the dentists was compared large differences were apparent. At one end of the spectrum, four dentists filled all carious primary molars, at the other extreme one dentist restored only 25 percent of carious molars and there was a gradual change between these two extremes. Following this study, the Oral Health Unit (OHU) of the National Primary Care R&D Centre recently completed a large qualitative study to gain a clearer understanding of how GDPs approach the care of young children.