Objective. Examine demographics, clinical characteristics and rehabilitation outcomes of lower-limb amputees, using the Australasian Rehabilitation Outcomes Centre (AROC) database. Methods. Lower-limb amputee rehabilitation separations between 2004 and 2010 were identified using AROC impairment codes 5.3-5.7.1 Analysis was conducted by year, impairment code, Australian National Sub-acute and Non-Acute Patient (AN-SNAP) classification (S2-224, Functional Independence Measure (FIM) motor(Mot) score 72-91; S2-225, FIM (Mot) score 14-71) and states of Australia. Results. Mean length of stay (LOS) for all lower-limb amputee episodes was 36.1 days (95% confidence interval (CI): 35.4-36.9). Majority of episodes were unilateral below knee (63.6%), males (71.8%) with a mean age of 67.9 years (95% CI: 67.6-68.3). Year-on-year analysis revealed a trend for increasing LOS and decreasing age. Analysis by impairment code demonstrated no significant difference in rehabilitation outcomes. Analysis by AN-SNAP found that LOS was 16.2 days longer for S2-225 than for S2-224 (95% CI: 14.7-17.8, P<0.001), and FIM (Mot) change was 12.0 points higher for S2-225 than for S2-224 (95% CI: 11.5-12.6, P<0.001). Analysis by states revealed significant variation in LOS, FIM (Mot) change and FIM (Mot) efficiency which may be associated with variations in organisation of rehabilitation services across states. Conclusion. Although amputees represented a comparatively small proportion of all rehabilitation episodes in Australia, their LOS was significant. Unlike many other rehabilitation conditions, there was no evidence of decreasing LOS over time. AN-SNAP classes were effective in distinguishing rehabilitation outcomes, and could potentially be used more effectively in planning rehabilitation programs. What is known about the topic? Literature reporting on the rehabilitation outcomes of cohorts of lower-limb amputees in Australia is limited to individual sites. No previous literature was identified that reported national data. What does this paper add? This study investigates amputee rehabilitation at a national level over a 7-year observation period (2004-10) and comprises 6588 episodes. It reports the national demographics, clinical characteristics and rehabilitation outcomes, with the aim of identifying findings that have implications for practitioners. What are the implications for practitioners? Although only a small proportion of all episodes in the AROC database, this subset of lower-limb amputee episodes has provided a useful snapshot of the current state of amputee rehabilitation in Australia. We believe these findings have significant implications for practitioners in delivery of amputee rehabilitation services across Australia. Practitioners may benefit from adjusting service delivery based upon the decreasing age of lower limb amputees. Findings from this study also indicate that AN-SNAP classifications are effective in discriminating amputee rehabilitation outcomes and may be used to streamline rehabilitation services and provide a more efficient and effective rehabilitation service to prevent further increases in LOS.