With Medicare’s growing focus on value-based payments, the debate is taking on new urgency
Quality metrics are assuming an ever-greater role in healthcare delivery, and many physicians are angry about it. They cite the vast increase in the number of metrics they must track, the lack of standardization among them, and the fact that payers are increasingly using them as a tool in setting compensation for medical practices and individual physicians.
By using metrics this way, doctors say, payers are downgrading the importance of doctors’ training and judgment in treating patients, and creating the potential for conflict between doing what the doctor feels is right for a patient and the doctor’s own financial well-being.
“I think [the growth of metrics] is a danger if it’s done wrong,” says Yul Ejnes, MD, MACP, an internist with Coastal Medical, a statewide internal medicine group in Rhode Island. “That’s why we need to use measures that are consistent with sound medical practice, as opposed to something someone thought was a good idea and didn’t get properly vetted.”
The proliferation of metrics, and providers’ growing unhappiness with them, are documented in recent studies. For example, a 2013 report by the healthcare consulting firm Bailit Health Purchasing LLC looked at 48 state and regional “measure sets”–such as the measures used by commercial payers for patient-centered medical home certification-and found a total of 1,367 individual measures. Moreover, Bailit found very little overlap, with only 20% of measures used in more than one such measure set.
In 2015 the Institute of Medicine issued its own report on the proliferation of quality measures, which found that “the growth in measurement and reporting activities results in considerable expense and requires substantial time commitment, without a matching return on investment.”
Also in 2015, in a joint study from The Commonwealth Fund and the Kaiser Family Foundation, 55% of the physicians surveyed said their practice gets incentives or reimbursements based on measure of quality of care, patient experience or efficiency of providing care. Meanwhile, 50% of physicians in the survey felt that the increased use of metrics to assess their performance has a negative impact on quality of care, with 22% saying the effect has been positive.
To help counter the problem of metric bloat, the Institute of Medicine has proposed a set of 15 standardized measurements ranging from life expectancy to patient safety to community engagement. (See box)
The growth of metrics is in some ways a consequence of the healthcare system’s ongoing transition from fee-for-service to value-based payments. Payers point out that if they are going to pay for improved outcomes and quality, they must have a way of measuring how well providers are doing at achieving these goals, and the ability to reward or penalize physicians and practices accordingly.
Thus, the ability of the two sides to balance the competing demands of measuring and rewarding quality in healthcare with patients’ individual needs will go a long way towards determining the success of value-based medicine.
The need to find that balance is about to assume greater urgency. Early last year the U.S. Department of Health and Human Services (HHS) announced a goal of having 30% of Medicare payments tied to quality or value by the end of this year, and 50% by the end of 2018. Moreover, the 2015 Medicare Access and Children’s Health Insurance Program Reauthorization Act (MACRA) will, beginning in 2019, further tie Medicare reimbursements to quality measures through its Merit-Based Incentive Payment System (MIPS) and incentive payments for participating in alternative payment models (APMs).
“The folks who are not doing well [in terms of quality] are going to be severely impacted by this legislation,” says Doral Jacobsen, MBA, FACMPE, senior manger with the accounting and consulting firm Dixon Hughes Goodman LLP. “And every year [after 2019] it will get a little worse.” On the other hand, Jacobsen notes, incentives in the legislation mean that a practice “could outperform their commercial contracts with Medicare if they are really rigorous in managing their costs and quality.”
To understand the dispute between providers and payers over metrics, it helps to look at metrics themselves and how they are developed. The Health Resources and Services Administration (HRSA) an arm of the U.S. Department of Health and Human Services, defines a quality metric as a tool that “allows a user to quantify the quality of a selected aspect of care by comparing it to a criterion.”
HRSA identifies five types of quality metrics, including:
For primary care physicians, probably the most important are patient outcomes. For these, most payers rely on some version of the Healthcare Effectiveness Data and Information Set (HEDIS) guidelines developed by the National Committee for Quality Assurance (NCQA). The 2016 HEDIS guidelines include 81 separate measures in five domains, (effectiveness of care, access/availability of care, experience of care, utilization and risk adjusted utilization, relative resource use, health plan descriptive information and measures using electronic clinical data systems.) More than half of the measures fall into the “effectiveness of care” category.
While there is no uniform method for developing a healthcare metric, most payers follow roughly the same process used in developing the HEDIS measures: selecting an outcome or aspect of care to evaluate, convening a panel of experts to review the pertinent scientific literature, developing the metric, vetting it with stakeholders, testing it with a limited number of providers, and rolling it out to all providers covered by the payer.
By now there is broad agreement in the medical community regarding the value of metrics in managing chronic diseases such as diabetes, asthma and congestive heart failure, says Joseph Scherger, MD, a family physician in La Quinta, California and member of the Medical Economics editorial advisory board.
“Those are the low-hanging fruit of chronic illnesses that, if controlled, will keep people out of the hospital and lower healthcare costs,” he says.
The current reliance on metrics had its beginning in the 1980s, when healthcare policymakers and payers began demanding greater accountability from providers, says Donald Berwick MD, MPP, former acting administrator of the Centers for Medicare & Medicaid Services (CMS) and author of numerous books and papers on healthcare quality.
“I’m all for transparency, but in my opinion we’ve overshot,” he says. “Any payer can ask for measurement and write their own rules as to what should be measured. It’s phenomenal, and it erodes productivity.”
John Meigs, Jr., MD, FAAFP, president of the American Academy of Family Physicians (AAFP) notes that the academy endorses the concept of value-based payments and thus the use of quality metrics, and his Centreville, Alabama, practice participates in a value-based incentive program through Blue Cross and Blue Shield of Alabama. Nevertheless, he adds, “as a practicing physician the quality measures at times are really frustrating, because I see what the patient needs, but I have to spend time because the insurance company or CMS wants to see certain numbers or certain boxes checked that don’t really help me take care of my patient.
“There needs to be some uniformity [in metrics],” Meigs adds. “It makes the data collection less intrusive, and it lowers doctors’ frustration levels.”
The lack of return on investment, experts say, usually occurs when metrics aren’t developed as part of a larger strategy. “What tends to happen is someone in authority says ‘we need measures,’ and lots of measures are produced, but they haven’t been incorporated into a framework for how they will lead to improvement,” says Eric Schneider, MD, MSc, senior vice president for policy and research at The Commonwealth Fund.
In such cases, Schneider says, metrics can even become counterproductive. For example, a metric might call for a certain percentage of patients to undergo colonoscopies. But without an upper age limit, “you run into the problem of doing them for patients who could be in the last year of life and have no chance of benefiting,” he says.
A further potential hazard posed by metrics is what Berwick calls “managing the measurement”-that is, focusing on the metric itself rather than on patient care. He cites his experience being in charge of quality of care management at a Boston HMO. “I had an unlimited budget for breast cancer screening and follow-up and we got good results for those. The problem was, as an oncologist pointed out to me, we really decreased the amount of work on breast cancer care. So measurement, by casting a light, led us to look under the light rather than thinking more broadly about the pattern of care.”
Of course, metrics can-and often do-play an important role in helping doctors, practices and institutions improve the quality of care they provide. But doing so requires planning, commitment and resources, says Ejnes of Coastal Medical.
Since its founding in the mid-1990s, Coastal, with 20 practices and more than 100 providers across the state, has invested in technology and quality improvement, including a single electronic health record system for all providers.
In 2015 Coastal, which has adopted the accountable care organization delivery model, was ranked among the top 1% for quality among practices participating in the Medicare Shared Savings Program.
Ejnes explains that each practice in the group gets regular reports summarizing its physicians’ progress in meeting a range of metrics, ranging from hemoglobin A1C levels for patients with diabetes to number of female patients undergoing mammograms. If the report shows, for example, that a patient is overdue for a mammogram, Ejnes’s medical assistant (MA) will contact the patient to remind her. If the patient is more than a year overdue, a standing order allows the MA to order one.
“A lot of the workflow for quality metrics involves people other than physicians, with the doctor being kind of the last resort if something can’t be rectified,” Ejnes says.
“Many of the docs here, myself included, used to cringe at this whole notion of getting a report saying these are your patients who haven’t had their cholesterol tested in ‘x’ number of months,” Ejnes says. “But between the fact that we now have data management resources and have adopted a team-based, we doctors now are in the position of managing [adherence to metrics] rather than implementing it.”
Coastal helps its providers meet the metrics required by different payers through the “Coastal Core”-a set of internal guidelines that, harmonizes differing payer goals associated with the same disease, such as A1C levels among diabetic patients. “Basically, we pick the most rigorous of the bunch so if we hit that we know we’ve hit everyone’s [goals],” Ejnes says.
A second, and perhaps more troubling challenge posed by metrics occurs when they are used as a basis for evaluating-and financially rewarding or penalizing-individual practices or physicians.
Metrics are developed using population statistics, but what’s deemed appropriate treatment across tens or hundreds of thousands of patients may not, in a doctor’s judgment, be right for the patient in front of him or her.
“I have tremendous doubts about the accuracy and utility and overall wisdom of quality metrics at the individual level,” Berwick says. “When it’s a ‘falling off a log’ standard that you know everyone would want the same way then it’s good to know we’re adhering to the standard. The problem comes when we don’t get a chance to modify them to fit individual needs. Can we really be smart enough to write rules that get down to the level of what’s right for each individual patient? So we have to be really careful about these rule-based standards.”
Another problem Berwick cites is that often doctors don’t provide care in isolation. “Physicians are enmeshed in a highly interdependent system of relationships,” he notes. “So maybe Dr. Jones’s patient had a particular outcome, but Dr. Jones is working with Dr. Smith and Dr. White. It’s hard to separate who’s doing what.”
A subtler, but still important effect of metrics as a reimbursement tool, Berwick points out, is to discourage doctors from caring for “difficult” patients-those with behavioral health disorders, or who are just plain ornery.
“If you put in a system that dings doctors for low scores on patient satisfaction or efficiency measures, but the problem lies in the variation of the patient population, you’ve done exactly the wrong thing,” Berwick says. “Instead of helping doctors take on the tough cases, you make it riskier for them.”
The possibility that metrics could cause a doctor to think twice before taking on a difficult patient is part of a larger and more delicate issue, which is the possibility of conflict between a physician’s judgment of what is best for an individual patient and the understandable desire to avoid being penalized or to qualify for a bonus by meeting a payer’s quality metric, or to get a high rating on Medicare’s “Physician Compare” website.
Medical societies’ ethical standards are unequivocal: the physician is obligated always to act in the patient’s best interest, regardless of the financial consequences, and few doubt that the overwhelming majority of doctors do so. Nevertheless, even the possibility of being placed in a situation where they must choose between doing what they feel is right for the patient and their own bottom line.
“As far as getting too fixated about meeting a threshold that isn’t reasonable and having that jeopardize what we do in the exam room and where our priorities lie, it’s something we’re always mindful of, and I talk about it constantly with my colleagues and medical director,” says Ejnes of Coastal Medical.
“Physicians’ judgment and knowledge of the patient should always override any particular measure,” says Meigs. “A lot of these measures are based on guidelines, and guidelines are suggestions, not edicts, although sometimes they look that way because the payer controls the payment and that makes them a little more forceful.”
“Doctors go out on a limb all the time, acting in what they think is the best interest of the patient,” says Scherger. “But you can see how this makes doctors angry and cynical. You’ve got to basically tell yourself that this [reliance on metrics] is probably a good thing overall, but when it’s applied on the ground it can be frustrating.”
Ultimately, experts say, the ability of metrics to improve the quality and delivery of healthcare depends on how they are developed and applied. “I would argue that quality metrics are very important, because we’re in the life and health business, so there needs to be a focus on quality control,” says Scherger. “On the other hand they should not be so onerous that they create a sense of burden on a medical practice, and they should not get so far out of perspective that the work against the more complex aspects of healing.”
“I think the jury is still out,” says Ejnes. He points to the example of a metric for lowering A1C levels for patients with diabetes to below 8%. “You can ramp up your practice to achieve that, but what do patients really care about? They care how many of your diabetics go blind or end up in dialysis. I’m not sure we’ve been able to connect those dots of the push to do measurements and those outcomes.”
For Meigs, metrics and the data that go into forming them are useful, but only as another piece of data to help diagnose and treat a patient. “You need that data just like we need all our medical knowledge and clinical judgment,” he says. “It all goes into me having conversations with my patients and trying to decide what’s best for them.”