In an initial study, Leake et al. (J Public Health Dent 1996; 52: 182-90) found that older Ontarians living in metropolitan communities received almost twice the amount of dental care than older Ontarians living in non-metropolitan communities. Since data were collected for the 2 years prior to the enrollment of subjects in a longitudinal study, factors responsible for this variation could not be explored. This paper extends these findings by examining the volume of dental services received by these subjects in the 3 years between the baseline and follow-up phases of the longitudinal study. During this 3-year period, geographic variations in the volume of services provided were also observed. In a regression analysis, geographic location remained a significant predictor of the number of services received after controlling for six other explanatory variables: insurance coverage, number of teeth, restorative treatment need, self-rated oral health, regular dental visits and use of specialist services. Together, these explained only 24% of the variance in service provision. Since patient-reported outcomes were better in the communities with higher volumes of provision, further research concerning the patient and dentist factors influencing treatment provision is warranted.