O’Keeffe et al (Reference O'Keeffe, Ramaiah and Nomani2007) highlight the lack of standards for evaluating liaison psychiatry services. They describe how timeliness of response is one possible quality indicator.

Our liaison psychiatry service serves a 600-bed general hospital in south London. Over a 3-month period in 2007 we audited our response times to 124 consecutive referrals against pre-existing standards.

We routinely categorise referrals according to the urgency of response required into one of the following three groups: emergency (including A&E) to be assessed within 1 hour; urgent, to be assessed within the same working day; routine, to be assessed within 2 working days.

For the three categories we achieved the response time standards for all referrals. The proportion of referrals in each group and the mean response times were as follows: emergency, 25%, 21 min (s.d.=20); urgent, 30%, 70 min (s.d.=86); routine, 45%, 200 min (s.d.=183).

A major advantage of an on-site liaison psychiatry service is the speed of response compared with psychiatric provision by community services (Royal College of Physicians & Royal College of Psychiatrists, 2003). it is difficult and expensive to conduct studies that might demonstrate the cost-effectiveness of a liaison psychiatry service in terms of reduced lengths of stay for general hospital in-patients and A&E patients. However, response time is a proxy measure. We hope that our data and the benchmarking recommendations by O’Keeffe et al will emphasise the importance of a high-quality liaison psychiatry service to healthcare commissioners.