Is now the time for the leaders of patient care organizations who are leading population health management initiatives to think broadly about some of the non-clinical aspects of health status, including food and nutrition, exercise, and other so-called “lifestyle choice” elements of health, in the context of care management efforts? Recent articles in The New England Journal of Medicine and Health Affairs certainly seem to point to that possibility. What’s more, a small number of patient care organizations are beginning to look at the sociodemographic and socioeconomic aspects of health status among their covered populations. Let’s look at what’s going on.
To begin with, I was fascinated to read a “Perspective” article that appeared online in The New England Journal of Medicine, entitled “U.S. Nutrition Assistance, 2018—Modifying SNAP to Promote Population Health.” The thought-piece, written by Sara N. Bleich, Ph.D., Eric B. Rimm, Sc.D., and Kelly D. Brownell, Ph.D., examines some of the policy foundations of one of the key federal anti-hunger programs in the United States, in the context of current population health management initiatives among U.S. healthcare providers and plans.
“The Supplemental Nutrition Assistance Program (SNAP),” the authors write, “is the cornerstone of the nutrition safety net in the United States, helping 45 million low-income Americans — nearly half of them children — pay for food each month. SNAP is authorized by Congress through the Farm Bill, which also covers agricultural programs such as crop insurance and land-conservation measures. With an annual cost of $74 billion, the program accounts for roughly 80 percent of the spending authorized by the bill. As an entitlement program, SNAP is responsive to economic fluctuations — enrollment can expand rapidly when the economy weakens and shrink when it improves. SNAP is scheduled to be reauthorized in the 2018 Farm Bill, which will set U.S. food policy for the next 5 years and beyond. As Congress deliberates, it’s important to consider what changes to the program are feasible and also have the potential to improve population health. Above all,” they state, “we believe SNAP should be protected — and, ideally, expanded, since its current benefits don’t allow most families to purchase adequate food to maintain a healthy diet.”
The researchers note that, while SNAP was never initially designed to focus on nutrition, but rather, was intended primarily to reduce hunger. Originally known as the Food Stamp Program, it was initiated in 1961 but didn’t become a permanent, nationwide program until 1974. SNAP has improved food security for millions of Americans. In 2014, SNAP lifted 4.7 million people, including 2.1 million children, out of poverty.” As the authors note, the challenge for many low-income families today “is less about obtaining enough food and more about finding dependable access to affordable healthy food. Currently, SNAP benefits can be used to purchase virtually any type of food or nonalcoholic beverage from eligible retailers.”
The authors reference a study that made use of point-of-sale transaction data from a leading grocery retailer that found that SNAP-recipient families “allotted a higher proportion of their grocery bills to soft drinks than to any other item (about 5 cents out of every dollar, as compared with 4 cents among non-SNAP households). It also found that both SNAP and non-SNAP households spent roughly 20 cents per dollar on sweetened beverages, desserts, salty snacks, candy, and sugar. Past studies involving nationally representative dietary-intake data have suggested that SNAP participants have poorer-quality diets than nonparticipants with similar incomes.”
Meanwhile, an article in the March issue of Health Affairs broadens out the subject. In “Impact Of The YMCA Of The USA Diabetes Prevention Program on Medicare Spending and Utilization,” authors Maria L. Alva, Thomas J. Hoerger, Ravikumar Jeyaraman, Peter Amico, and Lucia Rojas-Smith describe an innovative program devised by the YMCA of the USA, with support from a Health Care Innovation Award from the Centers for Medicare and Medicaid Services, that has been providing diabetes prevention education and coaching to Medicare beneficiaries with prediabetes, in 17 regional networks.
As the researchers note, “The YMCAs [participating in the program] use an evidence-based curriculum based on the Y’s adaptation of the National Diabetes Prevention Program of the Centers for Disease Control and Prevention (CDC). The goal of the Y model is to get participants to lose 5 percent or more of their body weight and gradually increase their physical activity to 150 minutes per week.”
The authors note that “The curriculum comprises 16 core sessions that cover the following topics: healthy eating strategies, understanding fat and calories, and elearning about foods that are high in nutritional value; strategies for increasing physical exercise, including incorporating exercise as part of one’s lifestyle and setting and achieving exercise goals; and strategies for changing one’s environment to help facilitate weight loss, using positive thinking, managing stress, and improving motivation. During the core sessions, lifestyle coaches facilitate group discussions of behavior changes, challenges, and solutions.”
The results speak for themselves: the researchers found that, comparing participants and nonparticipants, “[W]e found that the overall weighted average savings per member per quarter during the first rhree years of the intervention period was $278. Total decreases in inpatient admissions and emergency department (ED) visits were significant, with nine fewer inpatient stays and nine fewer ED visits per 1,000 participants per quarter. These results,” the researchers state, “justify continued support of the model.”
Drilling Down to the Data
Now, what does all this have to do with providers’ and plans’ population health initiatives? A fair amount, as it turns out. Just ask Justin Pestrue, administrative director of quality analytics at Michigan Medicine (the new name for the University of Michigan Health System), Ann Arbor, Michigan. On March 24, during the Health IT Summit in Cleveland, sponsored by this magazine, Pestrue presented “Effectively Using Sociodemographic Data in Healthcare Analytics,” a presentation that looked at sociodemographic and socioeconomic data, and the potential to leverage such data in efforts to move population health and accountable care initiatives forward.
As Pestrue told his audience last month, there is a very significant opportunity to leverage sociodemographic and socioeconomic data in new ways in order to more effectively care for patients in the context of accountable care organization (ACO) and population health work; and much of that opportunity has yet to be fully plumbed. As Pestrue noted, there is already a broad awareness of the fact that only a small proportion of patients’ health status can directly be impacted by care delivery in patient care organizations; some say that proportion accounts for perhaps 20 percent of the overall impacts on the health of individuals, with personal lifestyle and behavioral choices, personal environmental influences, and other influences being far stronger overall, he noted.
“[W]e have a real tendency to look at the data that’s easier to get—and that’s often the elements that are in our EHRs [electronic health records] and in our billing systems. But sociodemographic and socioeconomic data can do a really good job of helping us to better understand our patients,” he noted, with sociodemographic data include gender, race, ethnicity, age, and place of residence, and socioeconomic data including education level, income, etc., and both types of data being corralled together under the rubric “social determinants of health.”
In fact, in drilling down through the data they have, Pestrue noted that he and his colleagues are coming up with very interesting findings. For example, they’ve been drilling down several levels on location-related demographic data, scanning the service area around Ann Arbor by zip code. What they discovered was interesting, because zip code alone turned out to be a rather crude indicator of health status. Instead, drilling down to the sub-zip code level, they’ve found a variety of “hot spots” of neighborhoods of people with challenging sociodemographic and socioeconomic characteristics. And in one case, that work uncovered a sub-zip-code area—a neighborhood within the zip code in which Pestrue himself lives, which is a generally affluent area within the Ann Arbor metro area—in which the sociodemographic and socioeconomic characteristics of the population were far poorer, and in which ED visits and hospital admissions were higher. It turns out that that sub-zip-code area is a mobile home park.
And though he said that he and his colleagues have not yet moved forward to take action to identify individuals within that zone and health risk-assess them and get them into active care management, Pestrue told his audience that the potential is there to proactively intervene to improve the health status of such populations, using leading-edge analytics.
And clearly, there is potential here for a huge amount of progress in linking all these elements: looking at one’s covered population within any risk-based contract—whether an accountable care organization (ACO) contract with Medicare or with private payers, or involving any population health management initiative—and then linking the findings from any analytics work, into proactive engagement with patients who might be, as in the YMCA case study, prediabetic, to improve their health status before it worsens, through patient education, coaching, and care management.
Let’s be clear: the potential here is tremendous, absolutely tremendous. And as the public and private purchasers and payers of healthcare push the U.S. healthcare system further and further into value-based delivery and purchasing and further and further towards capitated or semi-capitated risk, bringing all these elements together will make economic as well as mission-based sense—really, it will.
So yes—the day is coming when patient care leaders will be diving far deeper on data and on linking that data to care management arrangements under ACO and population health contracts. And healthcare IT leaders will inevitably be drawn in very deeply on the analytics elements of this—as well as on the data architecture and interoperability efforts needed to connect it all together. Broccoli, anyone?