The hot topic with the patient community online, and at recent health conferences of every stripe, is the growing importance of the patient perspective in medicine. From trying to get doctors to listen to their patients better, to joint decision-making, much of this is long overdue.
But there’s a strange and alarming new trend on the extreme end of this subject that’s emerging rapidly, and that’s linking physician pay to patient satisfaction. Is this a good idea or a recipe for disaster? Well, before we can talk about that we need to review both the traditional way our doctors are paid, and the emerging new models of payment appearing on the scene since passage of the Affordable Health Care Act.
For decades, health services have been paid for using an à la carte menu. Insurance companies paid pre-negotiated fees based on each individual element of a visit. Suppose you went in for your quarterly diabetes check-up and annual flu shot. Your doctor would get paid a fee for how many minutes he or she spent with you (generally based on a metric of how complex the visit was), a second fee for your A1C test, a third fee for the fingerstick needed for that test, a fourth for the flu shot vaccine, a fifth for the syringe it went in, a sixth for the nurse that gave it, and so on. In the official parlance of the health-money world, à la carte is called fee-for-service.
The Slow Death of Fee-for-Service
But now, instead of this à la carte approach, most payors are looking at an all-you-can-eat buffet model. Physicians get a single flat payment for everything that happens at a visit. This is called a bundled payment. Why? Well, health plans don’t much like to part with their money and they suspected some docs might be stacking the deck with unnecessary charges. By developing a flat payment model for each type of visit, they argue that they are keeping costs down. Of course, many docs argue that the elements of visits that health plans define as necessary and appropriate (and thus reimbursed) are insanely inadequate, but that’s a story for another day.
One version of the bundled concept is an even broader disease management model in which a physician or practice gets a flat annual fee for keeping a patient healthy -- which opens up a new question: If, as an insurance company, you want to use that model, how do you know you are getting your money’s worth?
Pay for Performance
Thus the concept of fee-for-performance. Taking our restaurant analogy one step further, under this model the amount you’d pay for your meal would depend on how many stars the eatery gets from food critics. As you can imagine, this approach is highly controversial, with some good arguments to be made on both sides, but I’m personally against it.
Why? Well, frankly, I believe fee-for-performance puts an impossible burden on medical providers. It’s all good and fine to say we should pay only for success, but it’s not realistic. Human biology is too complex to guarantee success, and human behavior can torpedo even the best-laid plans of the smartest doctors. I cannot tell you how many times over the last decade I’ve seen diabetes patients go totally off the reservation and change their therapy without informing their doctors. The only way fee-for-performance could work is if we institutionalized everybody and forced them all to follow their therapy.
And as this would violate most of the articles of the U.S. Bill of Rights, I’m really not in favor of it.
Plus, I have other worries. Faced with losing money on hard-to-treat patients, will doctors discharge difficult patients from their care? Will chronically non-complaint patients, even if insured, be turned away from practice after practice like “narcotic seekers” are today?
But my distaste aside, if we adopt fee-for-performance as our defacto standard for healthcare pay, how do we measure performance? While some advocates for the system point to clinically measurable outcomes such as meeting benchmarks for A1C, blood pressure, and lipids, an increasingly vocal group of patients say that’s not enough, because it’s the patient experience that counts.
The Role of Patient-Reported Outcomes in Reimbursement
We, the patients, these folks say, need to have good real-life outcomes as well, not just good numbers. This includes therapies that are free from burdensome side effects, respectful interactions with healthcare providers, and quick and easy access to docs when problems arise. Welcome to the (possible) role of Patient-Reported Outcomes (PROs) in healthcare reimbursement.
The concept of PROs dates to 2008 with the creation of the PRO Consortium, a partnership among 26 of the largest pharma companies, the Food and Drug Administration, the European Medicines Agency (think FDA for the European Union), and the National Institutes of Health. Quoting from the Consortium’s website, their mission was to develop tools to measure the patient experience in clinical trials, that would look at, “one or more aspects of a patient’s health status based on information gathered directly from the patient, without interpretation by physicians or others. Patients provide information concerning the impact of an intervention or therapy from their perspective.”
So PROs started out as nothing more than finding a way to measure the patient voice in clinical trials. But putting science to that voice has turned it into a roar. PROs are now breaking out beyond clinical trials. They are starting to be incorporated into payor reimbursement models. In fact, if you’re on Medicare today and need surgery—and I hope you don’t—your satisfaction with the experience may well affect how much the government forks over to the hospital handling your procedure. I kid you not. Medicare’s new “value based purchasing” bonuses to hospitals are weighted 70% for clinical performance and 30% for patient experience.
How is the experience measured? Post-discharge surveys that ask patients about pain, room cleanliness, and whether or not the nurses and docs had attitude or treated them with respect.
What Does the Evidence Say?
But really, how relevant is the patient experience to the outcome? Is a happy patient always a healthy patient? Maybe not. A team of UC Davis researchers has found “that people who are the most satisfied with their (primary care) doctors are more likely to be hospitalized, accumulate more healthcare and drug expenditures, and have higher death rates than patients who are less satisfied with their care.”
The study’s lead author, Dr. Joshua Fenton, thinks that maybe being Mr. Nice Guy isn’t the best approach for doctors. OK, those were my words, not Dr. F’s. He actually said something more along the lines of, “Providers who are too concerned with patient satisfaction may also be unwilling to bring up uncomfortable issues such as smoking, substance abuse or mental health, which may then go unaddressed.” He also thinks patient-satisfaction-driven doctors may be more easily persuaded to order unneeded tests or procedures under patient pressure, and to minimize discussing risks in order to fulfill patient expectations.
This same study inspired leading social media doc KevinMD to write an impassioned editorial to put the kibosh on satisfaction-based pay. He thinks it’s bad medicine, saying, “We need more incentives to do less. Reward doctors for sticking to evidence-based clinical guidelines. Back them up for saying 'no' to patients, at the risk of lower satisfaction scores. Educate the public that more tests can, in fact, be harmful.”
He bluntly and strongly ends with, “How much more data do we need before realizing that patient care and patient satisfaction cannot be mixed?”
In the diabetes arena, I think that if we adopted a satisfaction-based payment model we’d quickly have no docs left to care for us. Yes, there is a long tradition in diabetes care of uncaring docs who did a poor job of thinking about the human side of the equation. But diabetes is tough and sometimes a tough guy needs to be in charge. After all, a patient who’s told by her doctor, “Sorry, you have to give up the 72-ounce Big Gulps” could well be unsatisfied by her medical visit.
But does that mean the doctor's pay should be docked for giving the advice?