April 22, 2025 • 16 mins
Let us know what you think about Health Affairs podcasts at communications@healthaffairs.org. If you have 30 minutes to spare, let us know and we'll set up a 30-minute chat for the first 20 listeners that reach out. Coffee will be on us.
Health Affairs' Senior Deputy Editor Rob Lott interviews Shuyue (Amy) Deng of Tufts University to discuss her recent paper that takes a closer look at the estimated impact of medically tailored meals on health care use and expenditures in the US.
Order the April 2025 issue of Health Affairs.
Currently, more than 70 percent of our content is freely available - and we'd like to keep it that way. With your support, we can continue to keep our digital publication Forefront and podcast
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Rob Lott (00:31):
Hello, and welcome to A Health Podicy. I'm your host,
Rob Lott. This April at HealthAffairs, we're all about food,
nutrition, and health. That'sthe subject of our new theme
(00:51):
issue, a veritable smorgasbordof important research, examining
the public investments andpolicies that really underpin
some of the most innovative dietand lifestyle interventions
aimed at things like improvingaccess to healthy, safe and
affordable food, eradicatinghunger and reducing health
(01:15):
disparities. One of thoseinnovative interventions is
medically tailored meals inwhich programs acting on a
referral of a medicalprofessional prepare and deliver
nutritionist designed meals topeople with complex health
conditions and high acuity care.
(01:37):
Such meals have been shown toimprove patients' health in some
cases, but we've also got a lotmore to learn about their impact
on healthcare use and spending.That's the subject of today's
Health Odyssey. I'm here withAmy Dung, a PhD candidate at
Tufts University's FriedmanSchool of Nutrition Science and
(01:59):
Policy. She's the lead author ona new paper from this month's
issue that attempts to envisionand quantify what the future
might hold for this kind ofintervention. Its title is,
quote, Estimated Impact ofMedically Tailored Meals on
Healthcare Use and Expendituresin 50 US States.
(02:21):
And so let's just dig in here.Amy Deng, welcome to A Health
Podicy.
Shuyue Deng (02:26):
Hi, thank you for having me.
Rob Lott (02:29):
Awesome. So let's just start with some background. Can
you describe a typical medicallytailored meal program? Who runs
something like this? What arethe typical patients and what's
their experience like?
Shuyue Deng (02:44):
So medically tailored meal programmes
typically are run by nonprofitorganisations, hospitals or
community based groups. Theyusually provide nutritionally
tailored meal designed byregistered dietitian and those
meals are specially provided forthe needs of individual with
(03:07):
diet sensitive chronic diseaselike diabetes, heart disease,
HIV cancer or kidney disease.Typically those patients are
like severely ill and oftenexperience food security and
unable to consistently accessthe nutritious food themselves.
Rob Lott (03:30):
And do you have a sense of how widespread this
intervention is? Can mostpatients who need it find one in
their neighbourhood?
Shuyue Deng (03:39):
So I think currently medical illiterate
meal program have expandconsiderably across The US
especially in recent year. And Ithink they are most established
in large urban centres likeplace in New York City, San
Francisco, Boston and LosAngeles. But availability is
(04:01):
lower in rural or less populousarea.
Rob Lott (04:04):
Do you have a sense of how effective medically tailored
meals are as an intervention?How good are they at reducing
healthcare utilisation orspending, improving people's
health? What's the evidence baseout there?
Shuyue Deng (04:22):
Previously a lot of study have shown medically
tailored meal programmes arehighly effective. They
consistently reduce hospitaladmission, emergency room
visits, healthcare costs and forthe coverage of high risk
(04:42):
population. For example,medically telomere intervention
have been associated with aboutsixteen percent fewer
hospitalizations. And how common
Rob Lott (04:54):
is it for something like medically tailored meals to
be covered by someone's healthinsurance policy?
Shuyue Deng (05:03):
So I think right now the health insurance are not
that commonly cover medicalilliterate meal but this
starting to change. Coverageusually happens through special
program like section elevenfifteen Medicaid waivers which
(05:24):
allows states to test theinnovative health approach like
medical retailer meals. And forexample, states like
Massachusetts, California andothers have already adopt those
waivers. And additionally, someMedicare Advantage plans and
(05:45):
private insurers are beginningto offer medically tailored
meals because they recognisetheir potential to save money
and improve patient health.
Rob Lott (05:54):
Okay, so against this backdrop, you conducted a study
with your co authors thatbasically asked, correct me if
I'm wrong here, what if everyonewho could benefit from or was
eligible to receive medicallytailored meals, what if they got
them and what did your studyfind?
Shuyue Deng (06:13):
So our research shown that if everyone who can
benefit from medically tailoredmeals actually receive them, we
find forty nine days have costsaving with Connecticut,
Pennsylvania and Massachusettsseeing the highest saving per
(06:34):
person. And only one stateAlabama showing cost neutral.
And we also find providing thismeal across US nationally can
save around 32,000,000,000 eachyear or to be processed in the
first year. For hospitalisation,we find providing the medical
(07:00):
entire meal in the first yearcan save around 3,500,000.0
hospitalisation each yearnationally.
Rob Lott (07:08):
Wow, that's a pretty big number. When you say
savings, just to clarify, you'rebasically comparing the cost of
implementing the program, payingfor the food, paying the
providers running the program,the less the savings from better
health, less cost expenditures.Is that a fair interpretation?
Shuyue Deng (07:31):
That's correct. That's the net cost saving,
which already subtract the feesfor like paying the meal and
dietitian screening.
Rob Lott (07:41):
And it sounds like you described there's a fair amount
of variation from state tostate. So in one state there's a
significant cost effectivenessin, I think you said it was
Alabama, it's less so. Whatexplains that discrepancy?
Shuyue Deng (07:57):
I think there are several drivers for the
variation. First, the eligiblepopulation size because states
differ in demographic and thenumber of people who qualify for
the programme. So that's onedriver and the other is the
(08:18):
baseline healthcare costs andthe hospital hospitalization
rate. In some states like wherethe healthcare is more
expensive, like Connecticut,Pennsylvania and Massachusetts,
there are more room to reducecosts and hospital use. And for
states like Alabama, thebaseline healthcare cost is not
(08:42):
as high as like in the states inthe Northeast Region.
In addition to those two,another I think the key driver
is the state specific healthcaresystem. Some states like for
example, Maryland have a uniquepayer payment model. For
(09:03):
example, Maryland's globalhospital budget system already
control costs in a differentway. So in Maryland, we can find
a different result.
Rob Lott (09:15):
Gotcha. So essentially in states where cost is maybe a
little more, you know, whereprices are higher, where cost is
not as under control, there's agreater potential savings from a
program like
Shuyue Deng (09:31):
this. Well,
Rob Lott (09:33):
I have a lot more to ask about the model. It sounds
really promising, but let'sfirst take a quick break. And
(09:59):
we're back. I'm here talkingwith Amy Deng, a PhD candidate
at Tufts Friedman School ofNutrition Science, all about a
paper in the April issue ofHealth Affairs in which she and
her co authors estimated theimpact of medically tailored
meals. Can you talk a little bitabout your timeframe as well in
(10:22):
the study?
I think you looked at outcomesafter one year and then again at
the five year mark. Why did youchoose those points and what did
you see in terms of a differencebetween those two points?
Shuyue Deng (10:38):
That's a great question. We choose to model
both one year and five yearoutcome to measure both the
short term and long term impactof the intervention. The one
year are mainly to show theimmediate effect like reducing
(10:59):
hospitalization and cost saving,which are important for decision
maker who want to see like aquick result. But we also look
at the five year outcome becausemany policies, especially at the
states or federal level areplanned and budgeted over
(11:19):
multiple year cycles oftenaround like five years. So
that's why we choose thistimeframe as well.
And we are not extending the earfor like even longer because
that may, I think compromise theaccuracy of our result.
Rob Lott (11:41):
Fair enough. Now this was a simulation which is sort
of the whole point, right, toimagine what's possible and
attempt to quantify it. But whatif a policymaker came to you,
for example, and said, this isall well and good. Now go ahead
(12:01):
and generate some real worldevidence. And let's also say
they maybe give you a blankcheck to fund that research.
Where would you start? What'snext? What are the questions in
this space that really haven'tyet been answered?
Shuyue Deng (12:18):
Yeah, thank you for this question. It's a great
question, because it's showingthe future potential research.
And I think the next steps islike because I understand the
simulation like ours help onlypaint the big picture of what
(12:39):
possible and but I doacknowledge like further real
world research can fill in thedetails and further guide the
implementation policy. If I havelike a blank check, I could
start with a large multi stagerandom control trial of
(13:01):
medically tailored meals. I willfollow patients over several
years and track healthcare useand spending and even include
different populations likeMedicaid, Advantage and eligible
(13:22):
individual to see how medcriteria meals perform in
different insurance andhealthcare system context.
And in addition, I would alsowant to look at who could
benefit the most and whetherthose effect differ by race,
(13:44):
disability status, differentdisability status and geography
and further more groups. So thenext steps, there are a few key
question I want to maybe askhere for future research
(14:05):
potential. For example, I wouldlike to see what is the optimal
dose of medically tailoredmeals. Like how many meals over
what period for what type ofconditions? That's one question.
And the answer is what happenswhen medically tailored meals
(14:30):
are paired with other supportlike nutrition counseling, SNAP
benefits or different carecoordination? So that's the two
questions I have in my mind butthere is a lot of more to
explore.
Rob Lott (14:48):
Wow, so that's a great agenda. So what's your sort of
final takeaway or your next bigstep?
Shuyue Deng (14:56):
The final takeaway is medically meal can prevent
hospitalisation and create costsaving for forty nine days. And
although we know medicallytailored meals work, but we want
to make more efforts to makethem more available, make this
(15:26):
intervention or medicallytailored meals become routine
part of healthcare. And to dothat we need more real world
evidence studies like randomcontrol trials to show how to
implement them well for peoplein need.
Rob Lott (15:46):
Well, that's a great agenda for the road ahead and a
great spot perhaps to wrap up.Amy Dung, thank you so much for
taking the time to chat with usand to tell us all about your
paper in the April issue ofHealth Affairs.
Shuyue Deng (16:01):
Yeah. Thank you. Thank you for having me.
Rob Lott (16:04):
It was it was great to talk to you. This was wonderful.
To our listeners, If you enjoyedthis episode, please tune in
again next week. Tell a friend,smash that subscribe button. And
until then, have a good week.
Thanks for listening. If youenjoyed today's episode, I hope
(16:27):
you'll tell a friend about ahealth policy.
Hello, and welcome to A Health Podicy. I'm your host,
Rob Lott. This April at HealthAffairs, we're all about food,
nutrition, and health. That'sthe subject of our new theme
(00:51):
issue, a veritable smorgasbordof important research, examining
the public investments andpolicies that really underpin
some of the most innovative dietand lifestyle interventions
aimed at things like improvingaccess to healthy, safe and
affordable food, eradicatinghunger and reducing health
(01:15):
disparities. One of thoseinnovative interventions is
medically tailored meals inwhich programs acting on a
referral of a medicalprofessional prepare and deliver
nutritionist designed meals topeople with complex health
conditions and high acuity care.
(01:37):
Such meals have been shown toimprove patients' health in some
cases, but we've also got a lotmore to learn about their impact
on healthcare use and spending.That's the subject of today's
Health Odyssey. I'm here withAmy Dung, a PhD candidate at
Tufts University's FriedmanSchool of Nutrition Science and
(01:59):
Policy. She's the lead author ona new paper from this month's
issue that attempts to envisionand quantify what the future
might hold for this kind ofintervention. Its title is,
quote, Estimated Impact ofMedically Tailored Meals on
Healthcare Use and Expendituresin 50 US States.
(02:21):
And so let's just dig in here.Amy Deng, welcome to A Health
Podicy.
Shuyue Deng (02:26):
Hi, thank you for having me.
Rob Lott (02:29):
Awesome. So let's just start with some background. Can
you describe a typical medicallytailored meal program? Who runs
something like this? What arethe typical patients and what's
their experience like?
Shuyue Deng (02:44):
So medically tailored meal programmes
typically are run by nonprofitorganisations, hospitals or
community based groups. Theyusually provide nutritionally
tailored meal designed byregistered dietitian and those
meals are specially provided forthe needs of individual with
(03:07):
diet sensitive chronic diseaselike diabetes, heart disease,
HIV cancer or kidney disease.Typically those patients are
like severely ill and oftenexperience food security and
unable to consistently accessthe nutritious food themselves.
Rob Lott (03:30):
And do you have a sense of how widespread this
intervention is? Can mostpatients who need it find one in
their neighbourhood?
Shuyue Deng (03:39):
So I think currently medical illiterate
meal program have expandconsiderably across The US
especially in recent year. And Ithink they are most established
in large urban centres likeplace in New York City, San
Francisco, Boston and LosAngeles. But availability is
(04:01):
lower in rural or less populousarea.
Rob Lott (04:04):
Do you have a sense of how effective medically tailored
meals are as an intervention?How good are they at reducing
healthcare utilisation orspending, improving people's
health? What's the evidence baseout there?
Shuyue Deng (04:22):
Previously a lot of study have shown medically
tailored meal programmes arehighly effective. They
consistently reduce hospitaladmission, emergency room
visits, healthcare costs and forthe coverage of high risk
(04:42):
population. For example,medically telomere intervention
have been associated with aboutsixteen percent fewer
hospitalizations. And how common
Rob Lott (04:54):
is it for something like medically tailored meals to
be covered by someone's healthinsurance policy?
Shuyue Deng (05:03):
So I think right now the health insurance are not
that commonly cover medicalilliterate meal but this
starting to change. Coverageusually happens through special
program like section elevenfifteen Medicaid waivers which
(05:24):
allows states to test theinnovative health approach like
medical retailer meals. And forexample, states like
Massachusetts, California andothers have already adopt those
waivers. And additionally, someMedicare Advantage plans and
(05:45):
private insurers are beginningto offer medically tailored
meals because they recognisetheir potential to save money
and improve patient health.
Rob Lott (05:54):
Okay, so against this backdrop, you conducted a study
with your co authors thatbasically asked, correct me if
I'm wrong here, what if everyonewho could benefit from or was
eligible to receive medicallytailored meals, what if they got
them and what did your studyfind?
Shuyue Deng (06:13):
So our research shown that if everyone who can
benefit from medically tailoredmeals actually receive them, we
find forty nine days have costsaving with Connecticut,
Pennsylvania and Massachusettsseeing the highest saving per
(06:34):
person. And only one stateAlabama showing cost neutral.
And we also find providing thismeal across US nationally can
save around 32,000,000,000 eachyear or to be processed in the
first year. For hospitalisation,we find providing the medical
(07:00):
entire meal in the first yearcan save around 3,500,000.0
hospitalisation each yearnationally.
Rob Lott (07:08):
Wow, that's a pretty big number. When you say
savings, just to clarify, you'rebasically comparing the cost of
implementing the program, payingfor the food, paying the
providers running the program,the less the savings from better
health, less cost expenditures.Is that a fair interpretation?
Shuyue Deng (07:31):
That's correct. That's the net cost saving,
which already subtract the feesfor like paying the meal and
dietitian screening.
Rob Lott (07:41):
And it sounds like you described there's a fair amount
of variation from state tostate. So in one state there's a
significant cost effectivenessin, I think you said it was
Alabama, it's less so. Whatexplains that discrepancy?
Shuyue Deng (07:57):
I think there are several drivers for the
variation. First, the eligiblepopulation size because states
differ in demographic and thenumber of people who qualify for
the programme. So that's onedriver and the other is the
(08:18):
baseline healthcare costs andthe hospital hospitalization
rate. In some states like wherethe healthcare is more
expensive, like Connecticut,Pennsylvania and Massachusetts,
there are more room to reducecosts and hospital use. And for
states like Alabama, thebaseline healthcare cost is not
(08:42):
as high as like in the states inthe Northeast Region.
In addition to those two,another I think the key driver
is the state specific healthcaresystem. Some states like for
example, Maryland have a uniquepayer payment model. For
(09:03):
example, Maryland's globalhospital budget system already
control costs in a differentway. So in Maryland, we can find
a different result.
Rob Lott (09:15):
Gotcha. So essentially in states where cost is maybe a
little more, you know, whereprices are higher, where cost is
not as under control, there's agreater potential savings from a
program like
Shuyue Deng (09:31):
this. Well,
Rob Lott (09:33):
I have a lot more to ask about the model. It sounds
really promising, but let'sfirst take a quick break. And
(09:59):
we're back. I'm here talkingwith Amy Deng, a PhD candidate
at Tufts Friedman School ofNutrition Science, all about a
paper in the April issue ofHealth Affairs in which she and
her co authors estimated theimpact of medically tailored
meals. Can you talk a little bitabout your timeframe as well in
(10:22):
the study?
I think you looked at outcomesafter one year and then again at
the five year mark. Why did youchoose those points and what did
you see in terms of a differencebetween those two points?
Shuyue Deng (10:38):
That's a great question. We choose to model
both one year and five yearoutcome to measure both the
short term and long term impactof the intervention. The one
year are mainly to show theimmediate effect like reducing
(10:59):
hospitalization and cost saving,which are important for decision
maker who want to see like aquick result. But we also look
at the five year outcome becausemany policies, especially at the
states or federal level areplanned and budgeted over
(11:19):
multiple year cycles oftenaround like five years. So
that's why we choose thistimeframe as well.
And we are not extending the earfor like even longer because
that may, I think compromise theaccuracy of our result.
Rob Lott (11:41):
Fair enough. Now this was a simulation which is sort
of the whole point, right, toimagine what's possible and
attempt to quantify it. But whatif a policymaker came to you,
for example, and said, this isall well and good. Now go ahead
(12:01):
and generate some real worldevidence. And let's also say
they maybe give you a blankcheck to fund that research.
Where would you start? What'snext? What are the questions in
this space that really haven'tyet been answered?
Shuyue Deng (12:18):
Yeah, thank you for this question. It's a great
question, because it's showingthe future potential research.
And I think the next steps islike because I understand the
simulation like ours help onlypaint the big picture of what
(12:39):
possible and but I doacknowledge like further real
world research can fill in thedetails and further guide the
implementation policy. If I havelike a blank check, I could
start with a large multi stagerandom control trial of
(13:01):
medically tailored meals. I willfollow patients over several
years and track healthcare useand spending and even include
different populations likeMedicaid, Advantage and eligible
(13:22):
individual to see how medcriteria meals perform in
different insurance andhealthcare system context.
And in addition, I would alsowant to look at who could
benefit the most and whetherthose effect differ by race,
(13:44):
disability status, differentdisability status and geography
and further more groups. So thenext steps, there are a few key
question I want to maybe askhere for future research
(14:05):
potential. For example, I wouldlike to see what is the optimal
dose of medically tailoredmeals. Like how many meals over
what period for what type ofconditions? That's one question.
And the answer is what happenswhen medically tailored meals
(14:30):
are paired with other supportlike nutrition counseling, SNAP
benefits or different carecoordination? So that's the two
questions I have in my mind butthere is a lot of more to
explore.
Rob Lott (14:48):
Wow, so that's a great agenda. So what's your sort of
final takeaway or your next bigstep?
Shuyue Deng (14:56):
The final takeaway is medically meal can prevent
hospitalisation and create costsaving for forty nine days. And
although we know medicallytailored meals work, but we want
to make more efforts to makethem more available, make this
(15:26):
intervention or medicallytailored meals become routine
part of healthcare. And to dothat we need more real world
evidence studies like randomcontrol trials to show how to
implement them well for peoplein need.
Rob Lott (15:46):
Well, that's a great agenda for the road ahead and a
great spot perhaps to wrap up.Amy Dung, thank you so much for
taking the time to chat with usand to tell us all about your
paper in the April issue ofHealth Affairs.
Shuyue Deng (16:01):
Yeah. Thank you. Thank you for having me.
Rob Lott (16:04):
It was it was great to talk to you. This was wonderful.
To our listeners, If you enjoyedthis episode, please tune in
again next week. Tell a friend,smash that subscribe button. And
until then, have a good week.
Thanks for listening. If youenjoyed today's episode, I hope
(16:27):
you'll tell a friend about ahealth policy.
Popular Podcasts
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.
Therapy Gecko
An unlicensed lizard psychologist travels the universe talking to strangers about absolutely nothing. TO CALL THE GECKO: follow me on https://www.twitch.tv/lyleforever to get a notification for when I am taking calls. I am usually live Mondays, Wednesdays, and Fridays but lately a lot of other times too. I am a gecko.