The American health care system has been plagued by poor medical outcomes and an inefficient cost structure for many years.1,2 The primary loser in this game is the patient, the player who bears the brunt of health care expenses, either directly or through taxes, and whose voice has been largely ignored until recently. As with many other service industries, health care delivery represents a highly complex process that is difficult to define or measure. Innovative procedures and state-of-the-art medical therapies have provided effective options that extend or improve life, but at increasing cost.
Further complicating the existing cost-vs.-benefit debate is a relatively new polemic regarding the degree to which patient satisfaction should be a measure of health care quality. How satisfied will a patient be with his or her health care experience? This depends, in large part, on the medical outcome.3
Customer feedback has long been considered a critical marketing and strategic planning tool for a variety of industries.4 Value, as defined by the consumer, is arguably the ultimate goal of any right-minded market competitor. In the digital age, customer questionnaires have become cheaper and easier to administer. Patient survey results can be used to help optimize the delivery of consumer-based value, measure employee performance, provide starting points for continuous improvement, and guide future strategy – all at a relatively low cost.
Many public and private organizations use patient satisfaction surveys to measure the performance of health care delivery systems. In many cases, patient satisfaction data are being used to determine insurance payouts, physician compensation, and institutional rankings.5 It may seem logical to adopt pay-for-performance strategies based on patient satisfaction surveys, but there is a fundamental flaw: Survey data are not always accurate.
It is not entirely clear what constitutes a positive or negative health care experience. Much of this depends on the expectations of the consumer.6 Those with low expectations may be delighted with mediocre performance. Those with inflated expectations may be disappointed even when provided excellent customer service.
Surveys are not durable. That is, when performed under distinct environmental conditions or at different times, surveys may not produce the same results when repeated by the same respondent.7 Surveys are easily manipulated by simple changes in wording or punctuation. Some specific encounters may be rated as “unsatisfactory” because of external factors, circumstances beyond the control of the health care provider.
Many health care providers feel that surveys are poor indicators of individual performance. Some critics highlight a paucity of data. A limited number of returned surveys, relative to the total number of encounters, may yield results that are not statistically significant. Increasing the amount of data decreases the risk that a sample set taken from the studied population is the result of sampling error alone. Nonetheless, sampling error is never completely eliminated, and it is not entirely clear to what degree statistical significance should be used to substantiate satisfaction, a subjective measure.
Surveys often provide data in a very small range, making ranking of facilities or providers difficult. For example, national polling services utilize surveys with thousands of respondents and the margin of error often exceeds plus or minus 3%. Data sets with a smaller number of responses have margins of error that are even greater. Even plus or minus 3% is a sizable deviation when considering that health care survey results often are compared and ranked based on a distribution of scores in a narrow response range. In the author’s experience, a 6% difference in survey scores can represent the difference between a ranking of “excellent” and “poor.”
Many patients are disenfranchised by survey methodology. In the most extreme example, deceased or severely disabled patients are unable to provide feedback. Patients transferred to other facilities and those who are lost to follow-up will be missed also. Many patient surveys may not be successfully retrieved from the homeless, the illiterate, minors, or those without phone or e-mail access. Because surveys are voluntarily submitted, the results may skew opinion toward a select group of outspoken customers who may not be representative of the general population.
The use of patient satisfaction surveys, especially when they are linked to employee compensation, may create a system of survey-based value. This is similar to the problem of defensive medicine, where providers perform medicine in a way that reduces legal risk. Aware that patients will be asked to fill out satisfaction surveys, associates may perform in a way that increases patient satisfaction scores at the expense of patient outcomes or the bottom line. Some institutions may inappropriately “cherry-pick” the easy-to-treat patient and transfer medically complex cases elsewhere.
It is not clear how to best measure the quality of a health care experience. With the broad range of patient encounter types and the inherent complexity of collaboration among providers, it is difficult to determine to what degree satisfaction can be attributed to individual providers or specific environmental factors. Patients do not typically interact with a specific provider, but are treated by a service delivery system, which often encompasses multiple players and multiple physical locations. Moreover, it is not always clear when the patient encounter begins and ends.
Despite the criticism of patient satisfaction survey methodology, the patient must ultimately define the value of the health care service offering. This “voice of the customer” approach is a diversion from the antiquated practitioner-centric model. Traditionally, patient appointment times and locations are decided by the availability and convenience of the provider. Many consumers have compensated for this inefficiency by accessing local emergency departments for nonurgent ambulatory care. Nonetheless,EDs often suffer from long waits and higher costs. Facilities designed for urgent, but nonemergent, care have attempted to address convenience issues but these facilities sacrifice continuity and specialization of care. In a truly patient-centric health care model, patients would be provided the care that they need, when and where they need it.
The patient satisfaction survey remains a primary tool for linking patient-centered value to health care reform. Ranking the results among market competitors can provide an incentive for improvement. Health care professionals are competitive by nature and the extrinsic motivation of quality rankings can be beneficial if well controlled. Employers should use caution when using survey data for performance measurement because survey data are subject to a variety of sources of bias or error. Patient survey data should be used to drive improvement, not to punish. Further research on patient survey methodology is needed to elucidate improved methods of bringing the voice of the patient to the forefront of health care reform.
References
1. N Engl J Med. 2003 Aug 21;349(8):768-75.
2. Centers for Medicare & Medicaid Services. Medicare hospital quality chartbook: Performance report on outcome measures. September 2014.
3. N Engl J Med. 2008 Oct 30;359(18):1921-31.
4. “Better Customer Insight – in Real Time,” by Emma K. Macdonald, Hugh N. Wilson, and Umut Konuş (Harvard Business Review, September 2012).
5. “The Dangers of Linking Pay to Customer Feedback,” by Rob Markey, (Harvard Business Review, Sept. 8, 2011).
6. “Health Care’s Service Fanatics,” by James I. Merlino and Ananth Raman, (Harvard Business Review, May 2013).
7. Trochim, WMK. Research Methods Knowledge Base .
Dr. Davis is a pediatric gastroenterologists at University of Florida Health, Gainesville. He has no financial relationships relevant to this article to disclose.
