ABSTRACT
BACKGROUND: The use of complementary therapies by patients has increased over the past 20 years, both in terms of self-medication and physician prescriptions. Among herbal medicines, those containing extracts of Asteraceae (Compositae), such as Echinacea spp., Arnica montana, Matricaria recutita and Calendula officinalis, are especially popular in the primary-care setting. However, there remains a gap between the growing acceptance of these remedies and the lack of data on their safety. OBJECTIVE: The aim of this study was to analyse prescribing patterns and adverse drug reactions (ADRs) for Asteraceae-containing remedies in Germany. METHODS: Primary-care physicians, all of whom were members of the German National Association of Anthroposophic Physicians were invited to participate in this prospective, multicentre, observational study. During the study period (September 2004 to September 2006), all prescriptions and suspected ADRs for both conventional and complementary therapies were documented using a web-based system. The study centre monitored all ADR reports and conducted a causality assessment according to Uppsala Monitoring Centre guidelines. Relative risks (RRs) and proportional reporting ratios (PRRs) were calculated. RESULTS: Thirty-eight physicians, 55% of whom were general practitioners and 45% were specialists, fulfilled the technical requirements and were included in the investigation. Because documenting all ADRs (i.e. serious and nonserious) was time consuming, only a subgroup consisting of seven physicians agreed to report nonserious in addition to serious ADRs. During the study period, a total of 50 115 patients were evaluated and 344 ADRs for conventional and complementary remedies were reported. Altogether, 18 830 patients (58.0% female, 60.3% children) received 42 378 Asteraceae-containing remedies. The most frequently prescribed Asteraceae was Matricaria recutita (23%), followed by Calendula officinalis (20%) and Arnica montana (20%). No serious ADRs for Asteraceae-containing remedies were reported. In the analysis of the subgroup of seven physicians who also documented nonserious ADRs, 11 nonserious ADRs for Asteraceae-containing remedies occurred in 6961 patients, resulting in an RR of 0.13 (95% CI 0.07, 0.23). The majority of reported ADRs for Asteraceae-containing remedies were classified as uncommon. A subgroup analysis comparing phytotherapeutic and homoeopathic preparations did not reveal any relevant differences. The PRR for Asteraceae-containing remedies with respect to all other prescriptions was 1.7 (95% CI 1.0, 2.0) for the system organ class 'skin and subcutaneous tissue disorders' (six ADRs) and 1.0 (95% CI 0.3, 3.6) for 'gastrointestinal disorders' (three ADRs). Neither result was significant according to the PRR criteria developed by Evans et al. CONCLUSION: This is the first study to provide a systematic overview of prescribing patterns and ADRs for Asteraceae-containing remedies in the German primary-care sector. Asteraceae-containing remedies were used frequently in this context, especially among children. Our results indicate that treatment with Asteraceae-containing remedies is not associated with a high risk of ADRs.