A study published in the March issue of Health Affairs confirms what many have suspected all along: medical practices that care for large numbers of high-needs patients—particularly larger-sized medical practices that do so—realize better patient outcomes. Those familiar with Lean management principles would say that of course this would be true. But having this documented and presented in Health Affairs is still extremely helpful.
The article, entitled “Outcomes For High-Needs Patients: Practices With a Higher Proportion of These Patients Have an Edge,” was written by a team of researchers—Dori A. Cross, Genna R. Cohen, Christy Harris Lemak, and Julia Adler-Milstein. The researchers, from the University of Michigan, the University of Alabama-Birmingham, and Mathematica Policy Research, write this: “High-value primary care for high-needs patients—those with multiple physical, mental, or behavioral health conditions—is critical to improving health system performance. However,” they note, “little is known about what types of physician practices perform best for high-needs patients. We examined two scale-related characteristics that could predict how well physician practices delivered care to this population: the proportion of patients in the practice that were high-needs and practice size (number of physicians). Using four years of data on commercially insured, high-needs patients in Michigan primary care practices,” they write, “we found lower spending and utilization among practices with a higher proportion of high needs patients (more than 10 percent of the practice’s panel) compared to practices with smaller proportions. Small practices (those with one or two physicians) had lower overall spending,” they note, “but not less utilization, compared to large practices.”
As the article’s authors note, “It can be challenging for physician practices to manage the needs of healthy and complex patients in parallel, because high-needs patients require more time and resources to treat than healthy patients, and rely more heavily on care management and social services that might be hard to integrate into routine care.”
In terms of the practices studied, the most common chronic conditions and other diseases involved were type 2 diabetes, chronic obstructive pulmonary disorder (COPD), liver disease, asthma, and cancer. In terms of the practices, 63 percent of practices had a minimal proportion of high-needs patients; 34 percent had a moderate proportion; and about 3 percent had a substantial proportion. Meanwhile, 71 percent of the practices studied were small.
As for outcomes, when practices with a minimal proportion of high-needs patients (less than 2 percent), patients in practices with moderate (2-10 percent) and substantial (greater than 10 percent) proportions, had lower spending across all measures. Total medical-surgical spending was nearly 12 percent lower for patients in moderate practices, and more than 40 percent lower for patients in substantial practices. Patients in moderate and substantial practices also had lower odds of incurring any inpatient spending, and, for practices with substantial high-needs patients, lower average total inpatient spending, as well as lower odds of incurring any ED spending.
These numbers are significant. Essentially, these medical researchers found, medical practices that have substantial percentages of high-needs patients—patients primarily with (often-multiple) chronic illnesses, as well as diseases like cancer that require intensive interventions—are seeing 40-percent lower total medical-surgical spending. And in this context, “substantial” practices are those with more than 10 percent of their patients being high-needs patients.
Now, it’s also true that smaller practices with a volume of high-needs patients saw 7 percent lower total medical-surgical spending. But the different between 7 and 40 percent is, of course, quite significant. And, despite the fact that “Patients in practice with a substantial proportion of high-needs patients had an average quality composite score nearly 5 percentage points worse than that of patients in minimal practices,” anyone who’s worked in an outpatient clinic setting can clearly see the benefit of more extensive, intensive delivery of care for high-needs patients.
There are some nuances here. For example, the authors write, “Our second set of findings regarding practice size revealed that small practices had lower spending, but not lower utilization, compared to large practices. The fact that spending and utilization were not both significantly lower might be explained if patients in small practices experienced a similar number of encounters, but with lower length-of-stay or intensity of needed services”—in other words, these providers “did a better job of maintain these patients’ health status such that they avoided serious declines that resulted in higher-intensity utilization.” The authors speculate that the potentially closer physician-patient relationships in smaller practices might be at play there.
Meanwhile, as the authors note, “Our findings were not uniformly positive in favor of practices with a higher proportion of high-needs patients and small practices, as both performed worse on the composite quality measure. It may be,” they speculate, “that for high-needs patients, adherence to evidence-based guidelines does not feel as pressing as keeping patients out of the ED and hospital. This is supported by our finding that the medication management subcomposite, an important domain of quality for high-needs patients, did not reflect worse performance.”
Still, there is a lot to ponder here, and as the authors state in the conclusion to the article, “Our findings should prompt policymakers to further vet the assumption that practice consolidation will result in better for patients who are most in need. Once the mechanisms underlying our findings are better understood,” they add, “policy makers can consider experimenting with models that feature practices with a higher proportion of high-needs patients and perhaps strengthen efforts to support small practices to help achieve high-value care for this population.”
What’s clear to me is that healthcare IT and data analytics will play a very important role in this, as will data governance and IT governance and management. Senior clinician and executive leaders in medical groups, and to the extent that those medical groups have IT professionals in their trenches, IT professionals, already will be needing to ramp up their outcomes measurement, data analytics, and quality reporting capabilities to participate in either MIPS (the Merit-Based Incentive Payment System), or in advanced payment models (APMs) under MACRA (the Medicare Access and CHIP Reauthorization Act of 2015). Why not take the opportunity to figure out where they’re doing particularly well with patients with chronic and other diseases, and in particular, why not look at patient volume, as a factor in this?
I believe that, as the leaders of medical groups begin to assess and weigh these issues, they will benefit tremendously from coming to understand where their gaps are, and also where their particular strengths are. And they can also figure out how to optimize both their clinical and their financial outcomes in areas key to their success with patients—and with payers. And, as usual, healthcare IT leaders can prove themselves heroes here—by helping to uncover what needs to be uncovered, and help their organizations move towards better outcomes of all kinds.
And in the meantime, studies like this one will continue to shed light on where the U.S. healthcare system as a whole needs to go to improve outpatient care in the clinic setting—an exceptionally important set of concerns going forward.