Abstract Background: Multi-item scales for monitoring alcohol withdrawal reactions have been used since the 1970s, and since 1985 we have used a modified version of the Clinical Institute Withdrawal Assessment (CIWA) in our general hospitals. This study was conducted to determine whether a shorter version of the scale would prove easier to use without loss of accuracy. Methods: A simultaneous ‘crossover’ clinical audit using two hospitals. The shortened scale was developed from the existing one, and had 10 items as opposed to the previous 18. The patients were followed throughout their course and the incidences of complication, the frequency of sedation, the delay in initiating monitoring and the ease of use were recorded. Results: There were 106 patients managed with the old scale and 96 with the new. The rate of complication was not different, being 16% in patients managed using the old scale and 14.5% using the new scale; the rates of sedation were 49 and 48%, respectively. Patients managed with the new scale had a shorter course with a median duration of 27.6 h compared with 40 h. The time from admission to first recording of a score was 5.4 h for the new scale and 4.8 h for the old, which is not a significant difference. Both scores were used according to instructions, but staff reported that the shortened scale was easier to use. Conclusions: We conclude that a shortened form of the CIWA alcohol withdrawal scale works as well as the original and is simple to use.