Objectives-To identify randomised controlled trials (RCTs) which evaluate aspects of pre-hospital care; to perform categorisation by theme; to compare the sensitivity and precision of the search databases.Data sources-August 1997 updates of MEDLINE and EMBASE databases, using the Datastar online system. Papers published in 1987 or later were included, with no language restrictions. Study selection-A trial was eligible for inclusion if it was judged, by two independent and blinded assessors, that participants followed up in the trial were definitely or possibly assigned prospectively to one of two or more alternative forms of healthcare with random allocation or a quasirandom method of allocation.Results-The Literature search retrieved 849 papers, of which 569 (67%) were in MEDLINE and 486 (57%) in EMBASE. Forty one (5%) were confirmed as reports of RCTs or quasi-RCTs, and the total number of individual trials was 38. Ten of these trials dealt with thrombolytic drugs; 14 were concerned with other drugs, 12 with equipment, and two with other interventions. Four trials were based on a sample size of more than 1000, and seven reported a statistically significant effect on mortality. All 41 papers were in EMBASE, and all but one were also in MEDLINE.Conclusions-Evidence based policy making with respect to the organisation of pre-hospital services cannot depend on RCTs. In the current relative absence of such evidence, practitioners and decision makers must use alternative information sources. A future review could examine a broader range of literature and be based on a wider search of published and unpublished material.