408 Why Do Seizure Patients Return to the Emergency Department?

Abstract

We noticed an increasing trend in the frequency of return visits and utilization of the emergency department (ED) by seizure patients who were recently treated and discharged from our ED. This study was conducted to determine predicting factors for seizure-related return visits in the ED. We used an observational retrospective cohort design, conducted in the setting of an inner city teaching hospital to study seizure patients visiting the ED from June 2008 to June 2010 who were over 18 years of age, had a primary ED diagnosis related to seizure (excluding trauma-related conditions) and were not hospitalized. Medical Records were reviewed to determine the following: age, sex, domicile status, number of seizure visits, interval days to presentation, prior psychiatric diagnosis, completion of follow visits, and whether or not their anti-epileptic drug level was therapeutic. An analysis was performed using Pearson chi-square test (2 sided) to determine the predicting factors for return visit group. ANOVA test analysis was used to compare the return group with a control group who did not return within 90 days of initial presentation. The return group showed a total of 218 visits occurred in 65 patients with the mean number visits per patient at 7.81 (range 2-28). The mean age was 44 years (range 20-69). Sixty-six percent of the patients were male and 34% were female. On 200 (91.7%) of the visits, patients were reported as domiciled and the rest 18 (8.3%) were not. The mean interval to presentation was 25.7 days (standard deviation of 23.39). Patients during 85 (39%) visits had a noted existing psychiatric condition and 130 (59.6%) did not, with mental retardation in 3 (1.4%) visits. Follow-up was recorded after 35 (16.5%) of visits and not completed after 128 (58.7%) of visits. A follow-up visit could not be verified in patients of 54 (24.8%) visits. Anti-epileptic drug levels reported for visits are as follows: therapeutic in 15 (6.9%) initial visits (IV) and 30 (13.8%) return visits; sub therapeutic in 80 (36.7%) initial visits and 127 (58.8%) return visits; and unknown in 123 (56.4%) initial visits and 61 (28%) return visits. Using Pearson Chi-square test, return visits directly correlated with the following: prior psychiatric diagnosis (p=0.00), sub therapeutic anti-epileptic druglevel on initial visit (p=0.00) and initial follow-up not completed (p=0.001). A total of 14 patients (1 visit each) were also identified from the records studied as a control group who did not return to ED within 90 days after their initial visit for seizure (and were not hospitalized). A comparison was made between this control group and the return group using ANOVA test analysis and the only statistically significant differences in the control group was shown in the areas of being domiciled (p=0.039) and having follow-up completed after the initial visit (p=0.002). The failure to complete follow-up was a strong predictor for return visits. In addition, the presence of a sub therapeutic anti-epileptic drug level on initial presentation, as well as a having a psychiatric condition, showed a correlation with return visits. This warrants further investigation into creating strategies for these seizure patients at the time they present to the ED for care, to improve compliance with their follow-up visits and adherence to seizure medications, as well as consideration to other resources such as social worker services and psychiatric intervention.