How many of you have ever been asked to complete a survey about your medical care? Normally, we think of surveys as providing useful feedback for a business and/or offering praise for a job well done.
In medicine, however, these surveys are all about the nuance.
Think about a patient who is court-ordered to be in a locked psychiatric unit. How would we expect her survey results to be?
How about another patient who had life-saving surgery but continues with chronic pain?
Another common instance is a patient who requires an emergency room for treatment, but he has to wait for hours and hours because the ER is responding to several incoming patients with trauma from a severe vehicle collision. The patient might not know about this, but he might reflect that his wait was too long.
Believe it or not patient satisfaction surveys are often misleading. Survey results do not always lead to better medical practices and do not mean that doctors or physicians with higher ratings are better. In the medical community many studies have been conducted that reflect why these surveys from companies like Press-Ganey are not effective and may even result in a higher risk of mortality.
Response Rates in a Satisfaction Surveys
In a patient satisfaction survey, the results are only as accurate as the amount of people who actually participate. Participation is a huge factor influencing the effectiveness of these surveys. If people do not respond, their experience is not documented. Researchers have known this for years and thus participation has become a problem as companies try to increase response rates.
In a study conducted by J Sitzia and N Wood [ http://intqhc.oxfordjournals.org/content/10/4/311.short ] in the late 1990s, researchers investigated the reported response rates in published studies in order to improve the response rates in patient satisfaction studies so results would be more accurate. The researchers found that the mean response rate was much higher in a face-to-face approach than those that required a response via mail. Data responses in person were collected at a rate of 76.9 percent while if the survey was done by mail the response rate was only 67 percent. We also know that people with an unfavorable experience are more likely to respond than those with an average or good experience, which can sometimes result in a sample bias leaning toward negative results.
In order for a patient satisfaction survey to have any importance, the results must be representative of as many patients as possible. However, if surveys strive to get a higher response at the cost of confusing the patient, the results will be skewed.
Articulating the Nuance in a Survey
Because patient satisfaction surveys are designed to get as many responses as possible, results have become skewed in favor of the healthcare providers and results may be meaningless. Since most surveys report high levels of satisfaction, the questions and the method that these surveys determine satisfaction have come under scrutiny. How can every patient be highly satisfied especially when every medical experience is drastically different from the last?
The study published in 1999 by Brian Williams, Joanne Coyle, and David Healy [ http://www.sciencedirect.com/science/article/pii/S0277953698002135 ] aimed to investigate why so many patients reported such high levels of satisfaction. The researchers conducted in-depth interviews with recent users of mental health services to discover the nuances of their answers to the satisfaction survey. From the 29 interviews, the researchers discovered that patients often report their medical experience in positive and negative terms. Results showed that because the survey questions were so complex, they elicited a positive response from the patient although their experience was actually negative. The patient satisfaction survey did not tell the whole story.
Customer Satisfaction Linked To Higher Mortality
In 2012 researchers at UC Davis discovered that patients that report higher levels of satisfaction on their surveys not only have higher death rates but are more likely to be hospitalized and accrue larger medical bills.
“Patient satisfaction is a widely emphasized indicator of health-care quality, but our study calls into question whether increased patient satisfaction, as currently measured and used, is a wise goal in and of itself,” said Joshua Fenton, lead author of the patient satisfaction study [ http://www.ucdmc.ucdavis.edu/publish/news/newsroom/6223 ].
and assistant professor in the UC Davis Department of Family and Community Medicine.
The study revealed that an overemphasis on patient satisfaction may result in unwanted health consequences. Researchers found that doctors ordered tests and medical procedures simply to satisfy their patients not for medical reasons. This puts patients at unnecessary risk and increases the cost of their hospital bills.
Because some doctors receive incentivized compensation for positive patient satisfaction survey results, these same doctors may be unwilling to go over potential adverse effects of tests and treatments patients request. They may also withhold warnings against deadly habits such as smoking or excessive alcohol consumption for fear of getting a lower satisfaction rating from their patient. When fear of a bad rating on a Press-Ganey or other survey overpowers the medical providers desire to help the patient, problems arise. In one system in which I consulted, patients learned to call the on-call, overnight doctor for refills on prescriptions and then threaten the doctor with a bad review if he/she did not refill a potentially-dangerous prescription that should be discussed with the regular, daytime doctor.
Although the patient may be more satisfied in the short term, his or her overall health might be compromised and in the worse case, his or her life may be at risk. The most striking results from the UC Davis researchers’ findings is that death rates were higher for more satisfied patients. Over a four-year period, for every 100 people who died in the least satisfied group, there were deaths of 126 people – a 26% increase – from the most satisfied group.
So, next time you get a survey from your doctor’s office. Think about some of the nuance in the evaluation and see if you can provide responses that are most helpful to the overall process and your health.