Hospitalized patients who require admission to residential care are often thought to make prolonged and inappropriate use of hospital resources. There are no Australian data on the factors that contribute to length of hospital stay for such patients. The aim of this study was to determine the timing of critical steps in discharge planning for hospitalized patients who need residential care. We prospectively audited 100 consecutive referrals to an Aged Care Assessment Team (ACAT) from one acute hospital in South Australia. Case notes were examined to determine the timings of critical events in discharge planning. We found 47% of patients were discharged to a nursing home, 16% to a hostel, 11% died, 10% returned home and 16% went to another facility. The average length of hospital stay was 27.2 days, and an average of 8.4 days elapsed before a decision to seek residential care was first recorded. A further 4.5 days elapsed before ACAT referral, 4.6 days before ACAT approval and 9.7 days before a residential care bed became available. We conclude that people admitted to our hospital from the community and who subsequently need residential care, spend 36% of their stay awaiting a residential care bed. Most of their hospital stay has elapsed before residential care is considered necessary and referral and approval processes have been activated. Strategies to reduce length of stay should perhaps focus on the earlier recognition of the need for residential care and accelerated referral and assessment processes. Earlier involvement by social work and occupational therapy should be considered.