Clinical Operations
Abstracts
Jesus Roa, MD
Orlando Health/Orlando Regional Medical Center
Disclosure Relationship(s): Nothing to disclose
Carolyn Holland, MD, MEd
University of Florida
Johnathan Kennedy, MD
Orlando Health/Orlando Regional Medical Center
Background: There is growing emphasis on patient satisfaction with care as a marker of value of health services and reimbursement. Hospitals currently employ private survey vendors such as Press Ganey to collect this data but samples are often small and may have limited validity. This study compared Press Ganey survey scores to prospectively collected “onsite” survey scores at ED discharge using the same questionnaire to evaluate the two sampling methods.
Methods: This prospective cross-sectional study enrolled a convenience sample of adult patients being discharged from the ED of a tertiary care center September 1 to October 31, 2018. Research assistants provided patients being discharged with an anonymous hardcopy of a 36-question survey and allowed them to complete the survey independently. Press Ganey data from the same period were obtained. Questions were rated on a 5-point Likert scale: very poor (VP), poor (P), fair (F), good (G) and very good (VG). Frequencies and Somers’ D statistic were calculated.
Results: There were 504 questionnaires completed onsite and 262 compiled by Press Ganey. For the question “What is your overall rating of care received?” onsite versus Press Ganey ratings respectively were VP in 0.4% vs 5.7%, P 0.4% vs 5.4%, F in 3.3% vs 10.7%, G in 12.1% vs 12.3%, and VG in 83.7% vs 65.9% (p<0.001). For question “What is the likelihood of your recommending our ED?” onsite ratings versus Press Ganey ratings respectively were VP in 0.6% vs 8.2%, P 0.8% vs 5.1%, F in 2.9% vs 8.6%, G in 11.8% vs 12.1%, and VG in 83.9% vs 66.1% (p<0.001). “Courtesy of the doctor” was VP in 0% vs 3.9%, P in 0% vs 0.8%, F in 2.3% vs 9.7%, G in 11.1% vs 19.7% and VG in 86.7% vs 66% in onsite vs Press Ganey respectively (p<0.001). “Keeping you informed about your treatment” was VP in 0% vs 5.1%, P in 0.4% vs 5.1%, F in 3.5% vs 9.7%, G in 9.7% vs 18.3%, and VG in 86.4% vs 61.9% respectively (p<0.001). “Doctor concern for your comfort” was VP in 0.2% vs 5.4%, P in 0.2% vs 4.3%, F in 3.5% vs 11.2%, G in 10.9% vs 17.4%, and VG in 85.2% vs 61.6% respectively (p<0.001). Similar patterns were seen across questions.
Conclusion: Responses from the onsite survey were consistently and significantly better than those conducted by Press Ganey. Onsite surveys may provide a more representative sample and real-time interpretation of patient experience.