We analysed the characteristics of all malpractice claims arising out of telephone calls to Swedish Healthcare Direct (SHD) during 2003-2010 (n = 33). The National Board of Health and Welfare's (NBHW) investigations describing the causes of the malpractice claims and the healthcare providers' reported measures were analysed using Qualitative Content Analysis. The original telephone calls themselves, which had been recorded, were analysed using the Roter Interaction Analysis System (RIAS). Among the 33 cases, 13 patients died and 12 were admitted to intensive care. Failure to listen to the caller (n = 12) was the most common reason for malpractice claims, and work-group discussion (n = 13) was the most common measure taken to prevent future re-occurrence. Male patients (n = 19) were in the majority, and females (n = 24) were the most common callers. The most common symptoms were abdominal (n = 11) and chest pain (n = 6). Telenurses followed up on caller understanding in six calls, and mainly used closed-ended questions. Despite the severity of these malpractice claims, the measures taken mainly addressed active failure, rather than the latent conditions. Third-party communication should be regarded as a risk. When callers make repeated contacts, telenurses need to re-evaluate their need for care.