OBJECTIVE: To investigate how the allocation of referrals for a community rehabilitation service to triage categories affects waiting time from referral to first appointment, and whether other factors also contribute to variance in waiting time. DESIGN: A prospective cohort study. SETTING: A multidisciplinary outpatient community rehabilitation program within a large metropolitan health service. PARTICIPANTS: Consecutive adult patients (N=379) commencing rehabilitation over a 3-month period. INTERVENTION: Allocation of referrals to a triage category of 1 (most urgent) to 4 (least urgent) by allied health clinicians guided by a written protocol. MAIN OUTCOME MEASURE: The primary outcome was waiting time from referral to service commencement. RESULTS: The small group of patients (4%) allocated to the most urgent category had significantly shorter mean waiting times than the other 3 categories (mean, 4.8d vs 19.6, 26.6, and 19.4d for categories 2, 3, and 4, respectively). Regression analysis indicated that approximately 11% of the variance in waiting time was accounted for by the triage categories. Site of treatment (home or center) and diagnosis also made small contributions (4% combined) to variance in waiting time. CONCLUSIONS: The triage process ensured rapid service for a small number of urgent referrals, but made little difference to the waiting time of the vast majority of patients. Given the resources required for triaging patients, the results of this study lead us to question the value of the triage system in this setting.