392 Disposition Destination Does Not Affect Patient Experience Ratings in Real-Time in an Urban Academic Emergency Department

Abstract

Patient experience surveys are becoming an increasing portion of merit-based incentive programs, alternative payment models and improved patient outcomes. As such, the ability to sustain “high” scores has gained prominence as many programs focus on broad initiatives to improve patient experience. These programs often lack specificity and use delayed survey results for assessment. The ability to target initiatives to patients in real time may provide a real innovation as timely discovery of patients at risk for low experience scores may allow for targeted service recovery efforts prior to the official survey ratings. This study sought to assess if real-time ratings of patient experience are affected by projected disposition destination in the emergency department. This study is a subset analysis of a single center, IRB-approved, pilot study to evaluate if emergency physicians can predict actual patient satisfaction scores in real time in an urban academic ED. The goal was to assess for any differences in patient experience based on projected disposition in real time. During randomly assigned time blocks, a research assistant performed anonymous tablet-based patient experience surveys on both physicians and patients as they presented to an urban academic emergency department. The survey consisted of several questions directed at the treating physician including projected disposition of the patient (likely admit, likely discharge or unknown at this time) and 4 questions directed to the patient designed to assess standard patient experience ratings of physician performance (“Courtesy of the Doctor,” “Degree to which doctor took time to listen to you,” “Degree to which doctor kept you informed of your care,” and “Doctor’s concern for comfort while treating you”), using a 5-point Likert scale from 1 (“very poor”)-5(“very good”). Participation was voluntary and confidential. Differences in mean satisfaction scores and percentage of “top box” scores for patients were stratified by disposition for t-test and Chi square analysis. Total n=62 for the full pilot study, with n=52 (19= admit vs 33=discharge) for this subset analysis (9 responses were removed as disposition was “unknown” and 1 patient refused participation for all questions and removed from analysis); overall 98% response rate. Patients’ means scores for physician performance on admitted vs discharged were compared (t-test): “Courtesy” (4.68 ± 0.11 vs 4.64 ± 0.10, p=.76), “Listen” (4.79 ± 0.10 vs 4.58 ± 0.13, p=.26), “Informed” (4.48 ± 0.14 vs 4.58 ± 0.12, p=0.6) and “Comfort” (mean =4.26 ± 0.20 vs 4.48 ± 0.13, p=0.34). Chi square values on the % top box scores on admitted vs discharged patients were: “Courtesy” (68.4% vs 69.7%, p=.92), “Listen” (79% vs 69.7%, p=.47) m “Informed” (52.6% vs 69.7%, p=.22), and “Comfort” (47.4% vs 60.6%, p=0.35). This study is a first to assess patient experience, in real time, as stratified by projected disposition. While this small analysis showed no significant differences between mean satisfaction and “top box” scores across experience ratings, patients likely to be admitted did have slightly higher mean scores with respect to listening, while simultaneously rating physicians lower on keeping patients informed and with respect to comfort for both mean and % “top box” scores. Expanding on this pilot study may provide insight into the needs and failings that prevent improvements in the patient experience in the emergency department.