The paper describes the type and frequency of adverse events and transicnt reactions following consultations with professional acupuncturists. In a postal survey, involving 1,848 professional acupuncturists, all of whom were members of the British Acupuncture Council and practising in the UK, details of adverse events and transient reactions following treatment were recorded on standardised self-report forms. A sample size of 30,000 treatments was sought, and piloting indicated that a four-week period was required. Practitioners also provided information on themselves, including age, sex, length of training and years of practice. A total of 574 practitioners responded. 31% of the total population. These practitioners reported on adverse events and transient reactions associated with 34,407 treatments. No serious adverse events were reported, where these were defined as requiring hospital admission, prolonging hospital stays, permanently disabling, or resulting in death (95% CI: 0 to 1.1 per 10,000 treatments). A total of 43 significant minor adverse events were reported, a rate of 1.3 per 1,000 treatments (95% CI: 0.9 to 1.7). These included severe nausea and actual fainting (12), unexpected, severe and prolonged aggravation of symptoms (7), prolonged and unacceptable pain and bruising (5) and psychological and emotional reactions (4). There were three avoidable events: two patients had needles left in by mistake, and one patient had moxa burns to the skin, also caused by practitioner error. The acupuncturists also recorded 10,920 mild transient reactions occurring in 5136 treatments. 15% (95% CI: 14.6 to 15.3) of the 34,407 total. In terms of local reactions, there were reports of mild bruising (1.7%), pain (1.2%) and bleeding (0.4%). Practitioners reported that patients experienced an aggravation of existing symptoms after 2.8% of treatments. The most common mild transient reactions to treatment were feeling relaxed (11.9%) and feeling energised (6.6%). In this prospective survey of 34,407 treatments, practitioners reported no serious adverse events. This conclusion was based on data collected from one in three members of the British Acupuncture Council. Given that the whole membership delivers between one and a half and two million treatments a year, this is important evidence on public health and safety. When compared with medication routinely prescribed in primary care, the results suggest that acupuncture is a relatively safe treatment modality.