Healthcare Quarterly
Abstract
Exploring "patient satisfaction" is intuitively appealing as one way to understand the patient experience and to help guide healthcare providers to improve healthcare. The cover article by McKim et al., entitled "Emergency Department Patient Satisfaction Survey in Alberta's Capital Health Region" published in your recent issue (Vol.10, No.1), does little to advance our understanding of how to use patient satisfaction metrics in this regard. Unfortunately, McKim and his colleagues are not alone - the healthcare literature is replete with other articles that leave us hungering for a better way to conduct similar research and to turn the research results into actionable improvement strategies.
Their survey of almost 800 emergency department patients found that, overall, 85% of patients were satisfied with their care. What is not mentioned is that 50 years of patient satisfaction research has found that this is always the case - in study after study between 80 and 90% of patients are "satisfied." The authors go to some length to stress the "unique" nature of the emergency department setting, but, again, research shows that regardless of care setting there is little or no statistical variation in overall patient satisfaction.
What to do? As almost everyone was satisfied, the authors were left to conduct a lengthy analysis of socio-demographic factors such as gender and age and their relationship to satisfaction. Not surprisingly, the correlations were weak. Again, decades of research has demonstrated that this is not a fruitful avenue of enquiry. While elderly patients appear to be more statistically satisfied with their care, this fact is not much help in determining how to improve care processes, whether in the emergency department or elsewhere.
We also take issue with the way the survey was designed. The survey was not conducted using a standardized instrument. There is no discussion of response rate, an important factor to know when deciding if the results are generalizable. The questions were vetted by an expert panel of providers, when there is good research evidence that patients should participate in the formulation of satisfaction questionnaires. Finally, the survey methodology excluded patients who "left without being seen." In other words, perhaps the ultimate measure of dissatisfaction was left out - many hospitals track and report "left without being seen" routinely as a measure of emergency department performance, and for good reason. If waits are long and communication is poor, patients may "vote with their feet," and "left without being seen" numbers rise.
Sitzia and Wood published an excellent review and analysis of patient satisfaction research in healthcare in 1997. They dissect in some detail the difficulties with traditional patient satisfaction research, including confusion among researchers over what we mean when we say we are measuring "satisfaction." The evidence is that most patients are grateful for receiving care in what is inevitably a stressful situation and they will only express dissatisfaction if a very negative event occurs. The authors suggest that a more fruitful approach may be to shift the focus to patient "dissatisfaction."
Qualitative methods can yield actionable information for healthcare improvement. Ironically, McKim et al. had rich clues in this regard from spontaneous comments offered by almost two-thirds of respondents to their survey. Not surprisingly these comments repeatedly covered such areas as wait times, poor communications and inadequate facilities - all dimensions that previous research shows are important in patients' assessments of the quality of their care. While many paragraphs were devoted to an analysis of socio-demographic aspects of the survey, these spontaneous comments were relegated to a single paragraph. The authors comment that "only generic improvement strategies can be launched" in the absence of further research. We would suggest that a handful of targeted improvement strategies in areas like wait times and communications would be far more effective than further research on socio-demographic factors.
Perhaps this type of survey is adequate if the goal is to reassure the public that patients are largely contented. However, if our goal is to actually identify areas to improve care and the patient experience, then studying satisfaction is insufficient. We are reminded of the story that everything looks like a nail when the only tool you have is a hammer. In this case the hammer is a continued reliance by McKim et al. and others on quantitative survey research evaluating patient satisfaction, when a combination of quantitative and qualitative methods focusing on specific dimensions identified by patients as important should be used to point to areas where improvements can actually be made to patient care and the patient experience, whether in emergency departments or elsewhere.
About the Author(s)
Dr. Nancy J. Martin is an Associate Clinical Professor in the Faculty of Health Sciences, McMaster University, and Director of Research and Performance Metrics at Grand River Hospital, Kitchener.
John Ronson is a founding partner of the Courtyard Group, an international company working with its clients to transform healthcare.
References
Sitzia, J. and N. Wood. 1997. "Patient Satisfaction: A Review of Issues and Concepts." Social Science and Medicine 45(12): 1829-43.
Comments
Be the first to comment on this!
Personal Subscriber? Sign In
Note: Please enter a display name. Your email address will not be publically displayed