Why am I asking a question that seems to have only one sensible answer? If you are a physician or other mainstream healthcare provider, I think you know where I’m headed. Yes, I want to talk about one of the most cringe-inducing, hot topics in today’s American medical system – patient satisfaction. I touched on this issue briefly in How to Kill a Medical Practice; the title of that piece should give you some idea of how I feel about the way medical groups currently approach patient satisfaction. Today, I’d like to dig a little deeper into one of the scourges of our daily practice of medicine.
Why Patient Satisfaction has Become Huge
Historically, there has not been much of a focus on customer service in the delivery of healthcare. This is partly because in the past, it mostly took care of itself. In the good old days, there was a much larger proportion of doctors working in small, private practices. The receptionist and clinical staff knew everyone by name. There weren’t as many insurance hassles, for the patient or the doctor. Costs were more reasonable. The internet wasn’t a thing yet, so patients looked to their doctors – instead of blogs* – for advice. In summary, society respected doctors, and it was generally assumed that doctors were doing their best to help people.
Over time, there has been massive consolidation of medical groups, to the point where the little guys with a shingle out front have become rare. Where there used to be dozens of players in a healthcare market, there may now be just a handful of behemoths, all competing for the same pool of patients. But these patients are savvier than in the past, and they (rightly) want to make sure they’re going to the best doctors. Many of these people also (rightly) see value in being treated like humans with feelings, which can be challenging when dealing with any large business – medical or otherwise.
The increasing competition for patients’ business has led to a palpable shift toward the consumerization of healthcare. There’s only one problem with that – patients are not customers, consumers, or clients. And they most definitely are not “always right.” The highest quality, most compassionate, most ethical doctoring involves giving patients what they need, even when it doesn’t coincide with what they think they need.
In addition to the financial pressures from increased competition for patients’ routine healthcare needs, hospitals are also feeling the pain. They are grappling with a 2012 provision in the Affordable Care Act that withholds a portion (2% in 2017) of total Medicare reimbursements for their inpatient care. 30% of this 2% is disbursed only to those hospitals that are deemed to have acceptable patient satisfaction scores.
There is an axiom in the medical insurance world that whatever Medicare is doing, commercial insurers will soon follow. What is not as widely appreciated is the impact the government’s whims can have on the operations of independent medical groups. In point of fact, many healthcare delivery organizations have followed the government’s lead with respect to withholding a portion of physicians’ salaries, to be disbursed only for good patient satisfaction scores. My organization has actually doubled down on the philosophy and piloted compensation schemes to withhold salary for both performance and patient satisfaction. Of course, because doctors are often financially illiterate, our administration succeeds in getting away with calling the payouts “bonuses” instead of “holdbacks.” I’m sure you can imagine how much that irks me.
You can see how patient satisfaction has become a flashpoint issue. Medical groups and hospitals want to be able to tout high scores they believe will attract more business and more government revenue. Therefore, they exert all kinds of pressure on their physicians to raise their satisfaction stats. Patients now have the power to appropriately punish bad doctors, but they can (and do) wield that power to also punish physicians who have patiently explained why their fatigue and weight gain is not due to the mythical condition of adrenal fatigue. They lambast doctors who attempt to decrease the incidence of drug-resistant bacteria by declining to prescribe antibiotics for a viral respiratory infection. And they downgrade doctors who try to prevent opioid-addiction by being stingy with narcotic prescriptions.
How do we Define Patient Satisfaction?
Many of you may think that it is perfectly reasonable to tie compensation to patient satisfaction and physician performance. On a conceptual level, even I agree with the philosophy. The problem is that nobody – I mean nobody – has yet figured out how to adequately define patient satisfaction or physician performance. Since physician performance is not the focus of today’s post, let’s table it. I’d like to move on to how patient satisfaction is being defined.
Earlier, I mentioned that a portion of a hospital’s Medicare reimbursements is withheld for patient satisfaction. So how does the government decide which hospitals will earn back the withheld money? They send out a survey to patients, of course! Specifically, they ask a random sample of patients to rate their recent hospital stay on the HCAHPS questionnaire. “H-caps” is the acronym for the easily rememberable Hospital Consumer Assessment of Healthcare Providers and Systems. My mouth feels like it’s full of marbles just typing that out.
CMS (Centers for Medicare and Medicaid Services) touts H-caps as “the first national, standardized, publicly reported survey of patients’ perspectives of hospital care.” Is there value in having a standardized assessment of such a thing? Probably. But the devil is in the details, of course. A 2015 Atlantic article describes one of the pitfalls in using the survey data:
For example, in a section about nurses, the survey asks, “During this hospital stay, after you pressed the call button, how often did you get help as soon as you wanted it?”
This question is misleading because it doesn’t specify whether the help was medically necessary. Patients have complained on the survey, which in previous incarnations included comments sections, about everything from “My roommate was dying all night and his breathing was very noisy” to “The hospital doesn’t have Splenda…” …An Oregon critical-care nurse had to argue with a patient who believed he was being mistreated because he didn’t get enough pastrami on his sandwich (he had recently had quadruple-bypass surgery). “Many patients have unrealistic expectations for their care and their outcomes,” the nurse said.
In my world of outpatient medicine, there are a few companies in the country that have assumed responsibility for administering patient satisfaction surveys and collating the data for the medical groups paying for the privilege. If you have received medical care in the US, you have almost certainly received one of these surveys by email or snail mail. The survey typically asks you to grade your care provider and other members of the team using criteria like friendliness, quality of explanations, demonstration of concern, etc. A 5-point scale is typical, ranging from very poor (1) to very good (5).
This all sounds straightforward and reasonable, right? Wait for it. In my group, results have been reported to us for years using the metric of our percentile for the highest mark of “very good.” According to the survey company – I have to take their word for this, as they do not share other groups’ raw data with us – we are graded against our specialty-matched peers from around the country. Apparently, most groups using this company have amazing patient satisfaction numbers, as the slope of the curve is extraordinarily steep as you move down from very good to good. Therefore, it is quite possible to have a majority of goods and very goods but land in the 30th to 40th percentile.
Unfortunately, my group maintains a laser-like focus on that “very good” percentile number, so the message doctors like me constantly receive is, “Why don’t patients like you? Why can’t you be a better doctor?” And trust me, there are a lot of doctors like me where I work.
The Problem with Patient Satisfaction Data
So what’s the counterargument? For starters, where I and many of my colleagues reside on the percentile curve does not jive with our subjective experience of how satisfied our patients are.
Seriously, HD? You’re always harping on how proud you are to be a card-carrying practicer of evidence-based medicine. Now, you’re going to argue that your subjective impression of how satisfied your patients are is more reliable than the hard data collected by the survey company?
I know this doesn’t look good, but bear with me while I explain myself. Then you can call me out. Almost every day, I have at least one patient who: asks me to reassure them I’m not leaving the organization; tells me I’m the only reason why they keep coming to my group for healthcare (they’re mostly frustrated that their PCPs keep leaving); or thanks me for explaining things better than any doctor they’ve ever seen. In addition to satisfying these patients, I also work hard to satisfy my colleagues who refer their patients to me. For many of the last 9 years, I have been named a “Top Doctor” by my regional peers.
I get it; you think you’re a good doctor. Great. Why do your patient satisfaction numbers suck, then?
First of all, I don’t think they suck. As far as I know, most people rate me at least “good.” But “good” doesn’t do me any good, because my percentile rank is based on “very good.” For me, good = F. When defining patient satisfaction as my group does, I think we instantly create a huge cohort of under-performing physicians.
Second, I tend to see a large number of the “worried well,” people who believe they have a hormonal etiology for their vague symptoms. Often, I’ve been set up for failure before these folks even step foot in my office, because their naturopath has already convinced them that they have multiple deficiencies and excesses. When I patiently explain that most of what they’ve been told is inaccurate and, in fact, their “hormones” are normal, how do you think that is received? Sometimes, people are thankful. More often, they’re very dissatisfied, which is not surprising. They typically enter my office with multiple “problems” that an Endocrinologist should be able to fix, but they leave my office with fewer diagnoses than they came with and no answers. I absolutely hate it when I have to answer the question, “But what do I do now?” with “I’m so sorry that I can’t be more helpful, but I don’t know. I would encourage you to sit down again with your primary care doctor and see if the two of you can come up with any other ideas.” This does not bode well for my satisfaction stats.
So how do you explain all the other doctors who have great satisfaction scores? Don’t they have the same challenges you do? How do they do it, and why can’t you just replicate what they do?
Funny you should ask that. Of course I’ve wondered the same thing. If I consider myself a great doctor who patients like, how can I reconcile that with the fact that 60% of my peers are better-liked? 1-2 years ago, I decided to dig into this by doing some reading. There are now numerous written accounts of how doctors with previously abysmal satisfaction scores swallowed their pride, implemented strategies to improve satisfaction (sit down, face the patient instead of the computer, don’t interrupt for the first few minutes, etc), and boosted their percentile up to the 80s or 90s.
I found that most of the strategies were things I was already doing, but I vowed to be more conscious of each step, to make sure I was doing everything recommended. There were a few tips that involved things I was not always doing (e.g. thanking the patient for coming in at the end of the visit), so I endeavored to find ways to implement these in ways that didn’t make me vomit (e.g. saying “it was good seeing you again” or similar, instead of thanking them for entrusting their healthcare to me).
After more than a year, I have moved the needle…not at all. In fact, I think my numbers actually got worse after some of the structural/logistical changes my organization has made (e.g. bad parking, poor check-in procedures, and other things over which I have no control).
To answer your earlier question, I really don’t know how my Endocrinology colleagues around the country are killing it when it comes to their patient satisfaction stats. I am truly baffled, so if any of you have a unique Rosetta Stone for patient satisfaction, please share it in the comments below.
Does Patient Satisfaction = Better Healthcare?
Now this is a fascinating question. I touched on this tangentially in Why Subspecialist Physicians Go Rogue – Part II. In that piece, I discussed the physician ego and how it fuels the desire to help people, even if that means inappropriately convincing oneself and the patient that the patient has a disease within the physician’s realm of expertise. Some of these rogue doctors have hundreds of patients in their panels – all spinning their wheels – all becoming pathologically dependent on these doctors for more and more advice…none of which helps.
One of the things I find intriguing about the above situation is that these patients can be incredibly satisfied with their doctor, even when they’re not getting better. I see patients who swear that Dr. Jones is the “only one who ever helped me,” even as they are in my office trying to figure out why they have all the same symptoms that predated seeing Dr. Jones. It is a disturbing example of cognitive dissonance, but easily explicable: Dr. Jones validates everything these patients say. It’s one thing to validate patients’ symptoms, reassuring them you believe that what they are experiencing is real. It’s another thing entirely to validate the straight lines patients draw from A to B when reaching their own (usually erroneous) conclusions about their vague medical problems.
While I do know some highly competent doctors with great satisfaction stats, the aforementioned “people-pleaser” doctors with excellent patient satisfaction scores are not physicians to whom I would send my family members. Lest you think I’m going off the rails here, I have data that supports my position. In March 2012, a landmark study was published in Archives of Internal Medicine. Joshua J. Fenton, MD, MPH et al reported their analysis of data from 50,000 adults, showing that the most satisfied patients:
- were 12% more likely to be hospitalized.
- had total healthcare expenditures and prescription drug expenditures that were both 9% higher.
- were 26% more likely to die.
In case you need additional convincing, the author of the Atlantic article I mentioned earlier did her own research and found similar data:
I examined Medicare’s provider data for thousands of hospitals—the data on every hospital in the country that the agency makes publicly available. I found the hospitals that perform worse than the national average in three or more categories measuring patient outcome. These are hospitals, in other words, where a higher number of patients than average will die, be unexpectedly readmitted to the hospital, or suffer serious complications. And yet two-thirds of those poorly performing hospitals scored higher than the national average on the key HCAHPS question; their patients reported that “YES, [they] would definitely recommend the hospital.”
That’s right, folks. Patients who think their care is great are often totally wrong. I’d like to reiterate a finding from Dr. Fenton’s Archives paper, because it’s just that important: the most satisfied patients were 26% more likely to die. This study has been criticized intensely over the years, as is any medical study that contradicts strongly-held, prevailing dogma. While there are other studies that suggest patient satisfaction is not a good indicator of quality, this one became the poster-child for vilification. This should not be surprising, as it landed when physicians like me were especially fed up with satisfaction surveys that didn’t seem to capture the quality care we were providing. So we began pushing back hard against our administrative overlords, citing Dr. Fenton’s study as evidence that we needed to be more thoughtful about the interpretation and application of patient satisfaction results.
But it’s hard to teach old dogma new tricks, especially in my organization. Nobody in administration wanted to slow the train down, citing other studies that show higher patient satisfaction is a good thing. I absolutely agree that a patient who hates her doctor is much less likely to follow through with the doctor’s recommendations, let alone return to see that physician. I really don’t need a study to tell me that, but sure enough, there are studies out there that do just that.
The problem here is that we have no way of teasing out the highly satisfied patients who are getting great care from the highly satisfied patients who are receiving terrible care. Further, we cannot break out the cohort of physicians with mediocre scores who are delivering better care than some of the people-pleasers at the top. So what the heck are we doing placing this much emphasis on satisfaction scores and tying compensation to them? I’ll tell you what we’re doing. We are guaranteeing that some of our best doctors will be told that they are failures. This will, in turn, accelerate physician burnout, as physicians who take great pride in the care they deliver will be utterly demoralized. Unable to move the needle on patient satisfaction, they will fry to a crisp, leave clinical medicine, and we will be left with even fewer great physicians in practice. You think I’m exaggerating? It’s already happening.
Should You Hate Your Doctor?
Let me point out that the Archives study was observational, as is most such research. It cannot prove causality – it can only show association. So I’m not saying that being a satisfied patient will kill you, nor am I stating the contrapositive that if you want to remain alive, you should hate your doctor.
I simply believe we should be significantly more thoughtful about how we interpret and use patient satisfaction scores. If you have the time and don’t mind the requirement to register (for free), Medscape published a fantastic interview with Dr. Fenton, the author of the Archives study. In it, he offers a finely nuanced discussion of the ongoing challenges we face with assessing patient satisfaction.
I will leave you with the final question and answer from that interview:
Medscape: What can physicians do if they are not receiving high satisfaction scores but believe that they are delivering quality care by other measures?…
Dr. Fenton: It depends on how you define “low” satisfaction scores. Physicians consistently scoring at the bottom 20th percentile of CAHPS may be repeatedly making some simple fundamental communication missteps that are causing recurrent problems in their encounters. Those physicians might benefit from simple interventions…
On the other hand, if a physician’s satisfaction scores are in the middle of the bell curve for his or her peers and this physician is doing his or her best to communicate with respect, empathy, and care, then we have no compelling evidence to force that physician to change [HD: italics are mine].
Translation for all medical groups bombarding their good physicians with futile “patient experience” initiatives: BACK. OFF.
*Yes, I am aware of the irony in complaining about people looking to blogs for medical advice…on my medical blog.
Are you a physician who struggles with how to improve patient satisfaction? What seems to work for you, if anything? Do you believe that the scores have value, or are we simply trying to raise an arbitrary metric to an even more arbitrary threshold number? As patients, what makes you most satisfied with your doctor? What makes you most likely to actually fill out the survey and rate your doctor 5/5? Comment below!
By reading this site and interacting with me in the Comments, you agree to abide by my Disclaimer.