Over the course of his career, Dr. Seth Berkowitz has met with patients much like one of his first – a 300-pound farmer in rural North Carolina with diabetes and heart trouble.
“His own diet was highly processed food, and he knew that was making his health worse,” Berkowitz says. “You’d talk with him and he’d be like, ‘Oh, I know what I need to be doing. It’s just not an affordable thing for me.’”
Berkowitz says as he encountered the problem over and over – in North Carolina, the Bay Area and Boston – an idea hardened in in his mind. Many of his sickest patients would happily focus on health, if they could just deal with their more pressing problems first.
So Berkowitz, an internist, has designed a pilot program at Massachusetts General Hospital, where he works now, that he thinks just might work for his diabetic patients.
“We’re working with an organization that delivers medically tailored meals to people in their houses,” he says. “Good quality food. It’s made from ingredients grown in the Boston area. They have their own test kitchen.”
Berkowitz bets if he gets them eating healthier foods, his patients will need less medication, will move around more and will be more productive.
He is one of a handful of researchers trying to improve health for the patients sometimes called the 5/50s; the 5 percent of patients who use 50 percent of the resources. The program has the potential to improve health and save money on hospital bills paid by Medicaid and, by extension, American taxpayers.
It’s a growing idea in a healthcare landscape where providers are increasingly rewarded based on patients’ outcomes, not just the number of services provided.
The idea sounds simple, right? Spend $100 for a week’s worth of meals as a way to head off the repeated trips to the hospital that can run $6,000 to $10,000 a visit. Save a lot of healthcare by spending a little bit more on social services.
But here’s the problem: Doctors don’t know who is going to end up in the hospital. They just can’t predict it well enough, so they might spend lots of money on meals for people where food isn’t the problem at all.
At a clinic run by Massachusetts General Hospital in Revere, not far from Berkowitz’s Boston office, plenty of patients seem like they could be a good fit for the meals program.
There’s 71-year-old Tom Sullivan, who weighed 250 pounds when he first showed up 10 years ago.
Today, he weighs 316 pounds.
“Whatever I eat is either sandwiches or microwave — garbage,” he says. “Your body can tell you if you are doing good or not. And it hasn’t gotten any better.”
And there’s 39-year-old Carrie Walsh, who knows what food she wants to buy, even if she can’t afford it.
“I just feel really sad because I want to take care of myself,” she says, her voice catching. “And if I had the financial means, I’d be able to be eating better.”
Berkowitz says he’s got 1,500 Walshes and Sullivans he could enroll in the program, but he’s only got the budget for 50 of them.
What makes finding the right 50 even harder, says Harvard health economist Kate Baicker, is having healthy food on hand might not be enough.
“Maybe they don’t have a place for food deliveries,” says Baicker. “Maybe they don’t have adequate cooking facilities.”
The stated goal of this work — improving health and saving money — hinges on lining up the right patient with the right program.
Allison Hamblin with the Center for Health Care Strategies in New Jersey says the danger there is that healthcare providers won’t get it right.
“We need to prove that we can economically justify this work as part of healthcare,” she says. “And until we can demonstrate that this is cost effective, it won’t be a mainstream activity.”
Poor targeting could sink a program financially. And right now, many providers doing this work don’t target at all, relying instead on referrals or signups through a first-come, first-served basis.
Doctors and researchers in Washington state, San Diego and Dallas are developing predictive models they believe will take the guesswork out and lead to the right matches. That encourages Dr. Clemens Hong of Massachusetts General, a lead researcher in this field, but he says people must understand this problem needs more than a big data solution.
“I think the perfect marriage is data and relationships,” he says.
Hong says a trusting relationship — one where the healthcare workers really know the patient — is essential. Remember, he says, this is a group of people whose lives are turned inside and out.
He says a patient at high risk might start taking medications for depression, diabetes or congestive heart failure and do well for a while. “Then Mom dies, right? And then they get depressed again and stop taking their medication,” he says. “Then here we are all over again.”
That brings us back to Berkowitz, who is also trying to marry data and relationships.
In his modest, windowless office, Berkowitz reads off a couple questions to help him find patients best suited for the meal program.
He’s created a 100-question survey, with questions like, “In the last three months, did you ever put off buying medication so that you would have money to buy food?” Or “I worried that my food would run out before I had money to buy more. Was that often true, sometimes true or never true?”
Crude as that may be, learning a patient’s complicated backstory may be the most effective way to move forward.
Even if cumbersome questionnaires are the best they’ve got, people in the trenches, like Revere clinic nurse practitioner Christine Goscila, say they’ll take it.
“You can see how frustrating it is for the patient,” she says. “Their weight steadily rises, and the insulin steadily rises. And it’s this vicious cycle that’s never stopped, because that one issue that could fix it all isn’t being addressed.”
Goscila says if it comes down to the providing meals or providing more insulin to help patients like Walsh or Sullivan, her money’s on the meals every time.
Click the media player below to hear about a program in Dallas that says it’s saved millions of dollars by pairing the right treatment with the right patient.
This story was reported with the support of the Dennis A. Hunt Fund for Health Journalism and the National Health Journalism Fellowship, programs of the USC Annenberg School of Journalism’s California Endowment Health Journalism Fellowships.
If you’re a member of your local public radio station, we thank you — because your support helps those stations keep programs like Marketplace on the air. But for Marketplace to continue to grow, we need additional investment from those who care most about what we do: superfans like you.
Your donation — as little as $5 — helps us create more content that matters to you and your community, and to reach more people where they are – whether that’s radio, podcasts or online.
When you contribute directly to Marketplace, you become a partner in that mission: someone who understands that when we all get smarter, everybody wins.