The effectiveness of an open referral electrocardiographic monitoring service in identifying an arrhythmogenic cause for syncope was evaluated. Over 5 years, 7,364 patients of all ages underwent ambulatory 24-hour electrocardiographic (Holter) monitoring using a 2-channel recorder. Of these, 1,512 (20.5%) were referred because of syncope. During monitoring, 15 patients had syncope and 7 of the episodes were related to an arrhythmia, usually ventricular tachycardia. Presyncope was reported in 241 patients, with a related arrhythmia in 24. Thus, an arrhythmia-related symptom that could be diagnostic was present in only 2% of the patients monitored. However, syncope or presyncope without an associated arrhythmia might be considered a negative diagnostic clue and occurred in 225 (15%). High-grade atrioventricular block was present in 15 and ventricular tachycardia in 116; only 6 (5%) reported associated symptoms. An age-related incremental increase in atrial and ventricular arrhythmias was found. In 415 of the 1,004 patients (41%) aged 60 years or more, arrhythmias that are conventionally associated with sinoatrial disease were recorded. Using stringent diagnostic criteria, the sick sinus or tachybradycardia syndrome was present in 33 (3%). Many older patients (70%) were taking drugs that could be arrhythmogenic, hypotensive or both. It is concluded that an open referral 24-hour ambulatory monitoring service rarely results in identifying relevant symptom-related arrhythmias in patients with syncope. It records many asymptomatic arrhythmias that can compound rather than resolve the diagnostic problem in older patients, because the data obtained could lead to unnecessary therapy. An iatrogenic cause for syncope should always be considered.