Abstract The purposes of this study were to analyse trends in hip fracture (HF) epidemiology over a 13-year period (1994–2007) in the Australian Capital Territory (ACT), to assess the potential impact of concurrent changes in hormone replacement therapy (HRT) and bisphosphonate use and to present a new prediction of HFs in Australia up to 2021. Annual sex- and age-specific incidence rates (per 100,000 population) were determined and standardized using the Australian 2006 population. The projected number of HFs was estimated by two models applying age- and sex-specific HF rates averaged for 2002–2006 (model 1) or continuously changing as observed in this period (model 2, Poisson regression) to the projected population. In 2006 compared to 2001, the population ≥ 60 years in the ACT increased by 19.7%. Over the last 5 years the average annual incidence HF rate compared to the previous 3-year period decreased in females ≥ 60 years of age by 28.3%. Between 2001 and 2006 the number of prescriptions for HRT dispensed in the ACT declined by 54.6, while the number of prescriptions for bisphosphonate increased by 245%, accompanied by a decline in standardized incidence of HF rates of 36.4%, mainly in women (42.1%). This represents an annual cost for bisphosphonates per one prevented HF, of $A45,250 or $A576 person/year. Compared to 2006 the total number of HFs in Australia according to model 1 will increase in 2011 by 20.1% and in 2021 by 58.8%, but according to model 2 will decrease by 15.5% in 2011 and 27.5% in 2021. Our data suggest that the previously predicted rising trend in HFs in elderly women reversed, but did not so for men. This was coincident with a significant fall in HRT use and increased prescribing of bisphosphonates, which is cost-effective. However caution should be used in attributing causation as this is an ecological study. If trends in HF observed in 2002–2006 continue, the absolute number of HFs in Australia in 2011–2021 will stabilise or decline (which is more likely), despite the rapid ageing of the population.