If implemented, evidence-based medicine (EBM) in general practice will improve health outcomes for patients. This paper examines the views of 60 Australian general practitioners about EBM. While 57% of respondents had a computer in their surgery, 15% had Internet access and only 3% had access to the Cochrane Library at work. The most commonly cited barrier to EBM was 'patient demand for treatment despite lack of evidence for effectiveness' (45%). The next three most highly rated barriers related to lack of time. For each of three tasks of EBM, namely searching for evidence, appraising evidence and discussing the implications of evidence with patients, lack of time was rated as a 'very important barrier' by significantly more participants than lack of skills (McNemar's tests: chi2(1) = 7.1, P = 0.008, chi2(1) = 14.0, P = 0.001 and chi2(1) = 9.0, P = 0.003, respectively). Preferred resources for EBM included clinical practice guidelines (rated as 'very useful' by 55%) and journals that summarize research evidence, for example Evidence-based Medicine (52%). Systematic reviews were considered 'very useful' by only 15% of respondents, consistent with our finding that 30% did not understand the term 'systematic review'. Furthermore, 43% did not understand 'meta-analysis'. A minority indicated they understood the terms 'relative risk' (23%), 'absolute risk' (28%) and 'number needed to treat' (15%) sufficiently to explain to others. Skills development is crucial to achieve EBM in general practice.