Patient satisfaction is a buzzword in health care and in Washington. If you Google “patient satisfaction,” you will get more than 7.4 million results.

The results range from commentary and scholarly articles to details about the HCAHPS (Hospital Consumer Assessment of Healthcare Providers and Systems) survey. Innumerable companies will help you “improve your patient satisfaction scores.”

This focus appears to be well deserved. If Americans are going to spend $2.3 trillion on health care (of which 31% represents inpatient care), then they should be satisfied too, right? Maybe not.

In a national study of patient satisfaction, health care utilization, expenditures, and mortality published by investigators from the University of California, Davis, higher patient satisfaction with doctors was associated with higher inpatient utilization, higher cost, and increased mortality, compared with less-satisfied patients (Arch. Intern. Med. 2012;172:405-11). The ultimate conclusion of this observational study is that we really don’t know how patient satisfaction is related to other aspects of high-quality care. It has certainly been suggested that patient satisfaction is directly linked to high-value care (that is, quality divided by cost = value), but this current study raises serious questions about that conclusion.

More important to this discussion is the plan to link hospital reimbursement and financial payments to doctors based on patient satisfaction scores as measured by the HCAHPS survey.

In light of the available evidence, is this a good thing? Probably not. Unfortunately, in an effort to rein in the growth of health care expenditures in this county, our government is experimenting on the health care system. In the Institute for Healthcare Improvement model, change is seen as inherently good whether you are moving forward or backward. If everything is changing (even in a willy-nilly fashion), a positive breakthrough is more likely to occur. That sounds nice on paper, but these changes can and predictably will have adverse consequences that may hurt our patients more than help.

I would like to share a personal story that highlights the complexity of patient satisfaction and high-value care. As an employee of the Cleveland Clinic, I had an opportunity last spring to reduce my health insurance premiums if I participated in a program that included routine preventive care, exercise, and adoption of healthy behaviors. As a 43-year-old man, I was active and otherwise healthy, and up to that point I had never required a regular doctor. I contacted a colleague to be “my doctor,” as I needed a statement from him if I were going to participate in this program. At this point, my satisfaction score was low; I didn’t want to see a doctor at all.

I made an appointment for an executive physical (a nice benefit as a member of the staff health plan) with my new doctor. The executive physical includes a flexible sigmoidoscopy. At the time, I figured if I were going to get the “works,” I might as well get a colonoscopy instead. I asked my new doctor if this was all right, even though I knew it wasn’t indicated according to the preventive health guidelines. My doctor (my friend and colleague) said this was fine.

My satisfaction level was going up.

I proceeded to have a colonoscopy last July. Much to my dismay and shock, I learned that I had asymptomatic but advanced colorectal cancer at the age of 43. I am not sure what my patient satisfaction was at that time, but I certainly was not happy.

Approximately a week later, I was in preoperative holding at the Cleveland Clinic waiting for my colon resection. One of the nurses asked me where I would like to go after the surgery: the “VIP floor” (known as the Founders Suites) or the colorectal floor. As a physician and a member of the professional staff at my hospital for over 15 years, I knew I would get greater expertise with the colorectal nurses and staff on the colorectal floor. I was able to make an informed choice to accept the double room on the colorectal floor (with all the associated smells), compared with the private and lavishly appointed “VIP” suite. Although I knew I made the right choice, my satisfaction level went down. I seriously doubt that a patient without medical training and perhaps inside knowledge would have made the same choice that I did.

Sure enough, the nurses on the colorectal floor (and my wife) were diligent in making sure I was getting out of bed, walking, and doing all the things necessary to get my injured colon to work properly again. This included a minimization of my pain medications. Yes, I was a grumpy patient. I hurt, and moving around just hurt more. I didn’t want to get up regularly. I didn’t want my deep vein thrombosis prophylaxis shots; those really hurt, too. At the time, my patient satisfaction

level was low. In retrospect, it was exactly what I needed to get the best outcome.

I am happy to report that I am back to my good health – but what about my patient satisfaction through this ordeal? I would have been “happier” if I’d gone to the VIP floor. If I’d told the nurses on that floor to “come back later” for my necessary walk, they would have acquiesced. The VIP nurses would likely have given me all the pain medication I wanted – regardless of the potential effect on my colon motility.

As a physician, I can reflect on my experience and recognize that doing what is easy for our patients is not necessarily in their best interest. To maximize value, we have to do what is right for them, and it will not always align with their overall satisfaction. Patient satisfaction is important, and we must continue to track and measure how well we interact with our patients. We must improve how we manage expectations. But it is a mistake to link patient satisfaction with hospital and physician reimbursement.

High-value care is not simply giving our patients what they want when they want it.

Dr. Michota is the director of academic affairs in the department of hospital medicine at the Cleveland Clinic.