The role of adverse drug reactions (ADRs) as a cause of hospital visits varies depending on the type of hospitals. Our aim was to determine the incidence of drug-related emergency department visits to a district hospital, and to identify the drugs and patient groups involved. All patient visits to the emergency department of a Finnish district hospital were evaluated prospectively for 6 months. The physician on duty and a clinical pharmacologist selected all possibly drug-related visits for further scrutinising. The causality assessment (drug-related or not) was judged according to WHO criteria, based on the patients' files, including laboratory and other data. Of the 7113 evaluated visits, 167 (2.3%) were "certainly" or "probably" drug-related; 102 (1.4% of all) were related to ADRs and 65 (0.9%) to intentional overdoses. The most common ADRs were gastrointestinal symptoms (n=17) caused by antibiotics, opioids, nonsteroidal antiinflammatory or cytostatic drugs. The International Classification of Disease (ICD-10) codes on patients' files were insensitive to disclose ADRs. The ADR patients were older (mean age 57 years) than the intentional overdose patients (38 years; P<0.001). Males predominated in the intentional overdose group (38 males, 27 females) but not in the ADR patients. The majority of intentional overdoses was caused by psychotropics. The ADRs lead to hospitalisation in a higher frequency (51%) than did the intentional overdoses (35%). In conclusion, the incidence of "certainly" or "probably" drug-related visits to the district hospital emergency room was relatively low. The ICH-10 codes on patients' files were found to be insensitive to disclose the ADRs, even when they lead to hospital admission, casting doubts on the usefulness of ICH codes alone in ADR evaluation.