OBJECTIVE: To evaluate agreement between independent clinician raters using a triage protocol to prioritize referrals for occupational therapy and physiotherapy within a community rehabilitation program. DESIGN: The priority category allocated to consecutive referrals by one of six clinicians in the referral office was compared with a second rating made by an independent occupational therapist, blinded to the initial priority rating. SETTING: A centralized referral office staffed by allied health and nursing professionals designed as a single point of access for sub acute and ambulatory services within a large metropolitan health network. PARTICIPANTS: 214 referrals for adults requiring community based occupational therapy or physiotherapy rehabilitation for orthopaedic, neurological or other conditions (such as falls or cardio-respiratory conditions). MAIN MEASURE: Agreement (weighted kappa = κ(w)) between the two ratings. RESULTS: Overall agreement was moderate (κ(w) = 0.60), but disagreement occurred in 30% of cases. Professional discipline of the raters did not affect agreement. Agreement varied between diagnostic subgroups, with significantly lower agreement for referrals for rehabilitation following elective orthopaedic surgery (κ(w) = 0.25) than the other categories combined. Differences in agreement were observed between the four triage categories, with the lowest observed agreement in the most urgent category. CONCLUSIONS: Clinicians in a centralized model of triage showed only moderate agreement when making decisions about client priority for community rehabilitation for occupational therapy and physiotherapy.