Revisitas no programadas en un servicio de urgencias de medicina hospitalario: incidencia y factores implicados

BACKGROUND: To known the revisit rate in an emergency department (ED), to define the clinical and epidemiological profile of revisited patients, and to identify influencing factors for revisits. PATIENTS AND METHODS: During one year period, we included all revisited patients returning to ED before 72 h of a previous discharge from medical unit of ED. As controls we included the next patient seen after every case being discharged. We compiled clinical and epidemiological data from both groups. For revisited patients, we identify the cause of the revisit, changes in diagnosis and/or treatment, diagnosis mistakes and final destination of the patient. RESULTS: We identified 406 revisits (revisit rate: 1.42%). Multivariate analysis disclosed, as positive predictive factors for revisit, and age over 60 years (p = 0.006), male sex (p = 0.02), visit performed at level 2 (severe diseases) (p = 0.02), initial assessment by junior resident (p = 0.01), number of complementary procedures higher than 2 (p = 0.01) and gastrointestinal disease as diagnosis after the first visit (p = 0.03). On the other hand, dermatologic symptoms as initial complaint and unspecific symptoms (p = 0.01) were negative predictors for revisit. In only 16% of cases, the revisit did not imply changes in the diagnosis or treatment. Revisits were due to disease-related factors in 34% of cases, physician-related factors in 33%, patients-related factors in 10%, system-related factors in 3% and there were no relationship with the previous visit in 15% (in 5% of cases the cause was unassessable). The diagnosis error most frequently seen was "nonspecific abdominal syndrome". Seventy six percent of revisited patients were admitted. CONCLUSIONS: The revisit rate in our ED is relatively low. Patients being revisited have well-defined clinical and epidemiological profile. The majority of revisited patients require to be admitted.