S1E16: AI, Equity, and Data Infrastructure in Safety-Net Care (ft. Renee Crichlow, Codman Square Health Center)
Healthy Uptime Podcast Renee Crichlow (Codman Square Health Center) _ Jordan Cooper (Rackspace)-20261105_120905-Meeting Recording
November 5, 2026, 5:09PM
25m 3s
Jordan Cooper started transcription
Jordan Cooper 0:03
Dr. Renee Critchlow, the former CMO of Codman Square Health Center. She is also a family medicine physician, lecturer at Harvard Chan School of Public Health, and the past president of the Society for Teachers of Family Medicine. Renee, thanks so much for joining us today.
Renee Crichlow MD 0:18
I appreciate it. Thanks, Jordan, for inviting me.
Jordan Cooper 0:21
Sure. As background for our listeners, Codman Square Health Center is a non-profit, federally qualified community health center, or FQHC, providing comprehensive primary care, behavioral health, dental care, and preventive services to underserved populations in Boston's Dorchester neighborhood with 42 physicians on staff. And I'll remind my listeners,
to FQHCs are outpatient centers. There are no inpatient beds, so a little different from our typical guests. So Renee, today we're going to be discussing the themes of AI equity and data infrastructure in a safety net care environment. So first question, just to kick us off,
Large academic medical centers are investing hundreds of millions of dollars, billions of dollars in AI, while community health centers operate on a comparatively modest budget. All right, we know this already, but are we approaching a future where wealthy patients and wealthy healthcare systems and commercial payers, as opposed to populations covered by Medicare and Medicaid, receive AI enhanced healthcare while the underserved population
fall further behind. What would have to change for AI to become the great equalizer instead of the great divider?
Renee Crichlow MD 1:31
Well, I think those are really good questions. And what I would say, Jordan, is we have to start thinking about the health care dollar as clinical and operational. So the operational dollar
for small clinics has a very large impact. Just even supplying the infrastructure for billing and engaging is sometimes up to 20% of our overall spend. Now, in a situation where, if you look at what AI could do,
Jordan Cooper 1:46
Mhm.
Mhm.
Mm.
Renee Crichlow MD 2:04
to help be a great equalizer, it would be about making the ability for us to deliver healthcare to be more cost-effective. The fact that I've spent 6% of my revenue just to have a CPA organization push our billing through, if I had the ability to use AI,
products in ways that could facilitate the reduction of our operational cost, I get to shift that over to clinical care. That's where the big hit can be. The challenge around that is...
Which AI product? The challenge around that is, do I have the IT infrastructure to even engage with implementing that? The challenge around that is not having the bandwidth or expertise within the smaller organizations to use the products that help facilitate more efficient care delivery.
Those things are where AI can really help us equalize as the smaller places. I've been very fortunate. I've been a full spectrum family physician for over 25 years, meaning both inpatient and outpatient, and I've been in academia the whole time. I've seen what the big organizations can do, and being a CMO at a smaller place,
Jordan Cooper 2:58
Mhm.
Renee Crichlow MD 3:18
You see the constraints that you have due to the bandwidth of your staff and the expertise of your organization, which does not have the clinical IT infrastructure that a larger organization can actually just buy.
Jordan Cooper 3:34
So I actually have a number of questions that you just kind of brought to mind. Obviously, FQHCs are safety net institutions that are largely supported, I'm imagining, by public dollars. I don't imagine many commercial payer populations are going to FQHCs. So if that's the case,
I'm wondering, and you said obviously that these FQHCs have, you know, limited budgets for infrastructure. So I'm wondering, these are all just variations on the theme what I'm about to say, you know, why don't FQHCs band together? Why does the purchasing power of CMS not filter down to all FQHCs, all the BCBSs
band together for BCBS Association, why don't you partner with Boston Medical Center or with Tufts or with Bay State or what, you know, why aren't you leveraging your purchasing power? Why isn't FQHC needing the same thing as an FQHC in Kansas and Boston, all saying, here's a collective health information exchange, we're going to.
you know, put our data centers together and then we're going to have a massive amount of buying power because a few dollars here and a few dollars there collectively is a lot. Why doesn't that happen?
Renee Crichlow MD 4:44
Well, I think you hit the nail on the head. I feel a kindred spirit with you, Jordan. This is exactly the kind of thing that I've been trying to do in Massachusetts. We have an organization in Massachusetts called the Mass League of Community Health Centers, and it has 52 health centers around Massachusetts. One of the big challenges that we run into
Jordan Cooper 4:49
Okay.
Renee Crichlow MD 5:05
at larger scale, like federal, as you may understand, our most recent federal government has not been as...
as consistent in its ability to support marginalized communities. And so we, at the state level, we're looking to do exactly what you said. It's like, we've got 52 health centers, we can have a centralized purpose for vetting of products, we can have a centralized purpose for engaging with contracts.
Jordan Cooper 5:16
Supportive.
Mhm.
Renee Crichlow MD 5:37
And that is something that is on the table. Now that there's been probably a sea change in what the framework of what the infrastructure of our federal government is going to look like in the next couple of years, there may be opportunities. And when you think about FQHCs, what if they had the buying power that
Let's say the VA has. Let's, you know, using the idea, VA has a centralized system that's across the entire country. When you're looking at, the problem is, one of the problems is everyone says when you've seen one FQHC, you've seen one FQHC.
Jordan Cooper 6:00
Mhm.
Renee Crichlow MD 6:14
Everyone has a definite feeling of their specialness, which is rich and true. And yet we have a lot of the same problems. So approaching it at an economies of scale is something that a lot of us are very interested in doing. We are, Codman itself was partnered deeply with BMC.
Jordan Cooper 6:34
Mm-hmm.
Renee Crichlow MD 6:34
And using, we're on their EHR, Cotman uses a lot of the systems they have, but one of the challenges we run into is we might see something that they don't see a need for. So I'll give an example.
Jordan Cooper 6:47
Mhm.
Renee Crichlow MD 6:51
For people who have chronic lung illnesses, one of the things that we do on a regular basis for them is do pulmonary function tests. And pulmonary function tests at a tertiary medical center are big, huge tests. You test every single part of it. You do all the things. At our community health center,
Jordan Cooper 6:58
Mhm.
Mhm.
Renee Crichlow MD 7:12
We just need the spirometry part. I won't get into details of it, but it's a much simpler thing. And it doesn't need a pulmonologist. We need a machine. You blow into it. We have the ability to read out the information that comes out of it and make a diagnosis on whether this person has a particular type of chronic pulmonary
Jordan Cooper 7:31
Mhm.
Renee Crichlow MD 7:32
lung disease. So we at Codman were like, hey, we need our spirometry. We need to increase the number that we do. One of the ways you can do it is AI-facilitated diagnostics. It just pushes the product along very quickly, keeps the human in the loop. The physician at the end reads out their
Jordan Cooper 7:33
Mhm.
Renee Crichlow MD 7:51
the output and makes a decision very quickly. So it lowers the activation energy for getting this test done. So we went to our partners and we said, we need this AI product put into the system so that we can use it. And they're like, whoa, hold on. How's that going to affect us? And
Jordan Cooper 7:52
Mhm.
Mhm.
Renee Crichlow MD 8:10
We're like, it doesn't affect you at all. It helps us facilitate access for our patients to get better care.
Jordan Cooper 8:16
Mhm.
Renee Crichlow MD 8:17
And in order to do it, we went through months and months and months and months of both push and pull and push and pull. The challenge was the hospitals have a different...
They have a different set of priorities. So a big part of what we had to do was let them, one, understand that our priorities did not interfere with theirs. Our priorities were supportive of theirs. And the AI product we were looking to bring in was not going to in any way interface with them.
In doing that, they started reviewing the product and they were like, hey, maybe we should get this product too. And one of the things we found is as we approached the company, the more of the need we had, the lower the price would have been overall. So it actually worked out well for
Jordan Cooper 9:03
So...
Renee Crichlow MD 9:05
both of us, but we had to help them understand the value of it.
Jordan Cooper 9:06
Mhm.
I have a number of questions off that. The first one is, are they, were they worried, was Boston Medical Center initially concerned that there would be a reduction of patient volume for these complex testings, but then wouldn't, isn't the natural kind of response? Well, these are
Renee Crichlow MD 9:23
Yeah.
Jordan Cooper 9:29
indigent populations who are just, the alternative isn't going to BMC and getting a whole battery of tests. The alternative is just staying at home and getting no tests and no diagnosis.
Renee Crichlow MD 9:38
Yeah, so it's no tests and no diagnosis. And also, what we're looking to do is make sure the people that get sent to the tertiary care centers are the people that need to be in the tertiary care centers. So if they're like, oh, we have a backlog and we can't do our pulmonary function test because we have such a backlog, their access is so low. If you say,
Jordan Cooper 9:49
Mhm.
Mhm.
Renee Crichlow MD 10:09
people are going to be crowded out. So what we're actually doing is facilitating the access for our patients in their neighborhood, and we're reducing the backlog that would be at the medical center for people who don't need that level of care. And that's one of the ways that we had to try to explain it to people.
Jordan Cooper 10:11
Mm.
Mhm.
Renee Crichlow MD 10:28
Number one, our patients are getting back better access and your patients are getting better access because we're not throwing a bunch of folks in there that could be cared for in our environment.
Jordan Cooper 10:38
So it improves access. Let me ask you back on the previous topic, what would it take for FQHCs to band together, for example, in the state of Massachusetts, or just because this is a broader audience on this podcast in Virginia or California or Colorado or wherever?
would it take for, you know, both legally, politically, you know, clinically, operationally, right, like to convince that you're a special snowflake FQHC number one, but if you band together, you'll be able to get a whole bunch more access to budget than you would if you were on your own.
Renee Crichlow MD 11:03
The.
I would tell you that it would be interesting. What would be really helpful is if industry said, hey, let's approach these folks and tell them, FQHC1, if you get...
20 more FQHCs that are interested in this product, we can bring it down to a price that is going to be usable for all of you. So it would be really helpful if they had some outside information on that. The challenge coming from us is bandwidth and expertise. All the people that are working in FQHCs are doing 7 jobs.
Jordan Cooper 11:34
Mhm.
Mhm.
Mhm.
Renee Crichlow MD 11:49
And so the organizing around this has to have some real focus. And part of what the biggest for the FQHC is to do it, they would have to have dedicated people that would be able to organize this. For the industry to approach and say, hey, this is a good idea and we could be helpful for you.
Jordan Cooper 11:55
Mhm.
Renee Crichlow MD 12:09
That would help facilitate, reduce the activation level on that. We could get it done quicker if industry recognized that it was worthwhile for them. It's not worthwhile for them to sell to one. It's worthwhile for them to sell to 50.
Jordan Cooper 12:20
So, if you make...
Right. All right. Interesting. So the, I guess what James Carville said in 1992 about the Clinton campaign, it's the economy is stupid. When you speak to people's pocketbooks, then that's what gets their attention. Let's transition to a new topic. So community health centers possess something enormously valuable, which is longitudinal patient data on populations
Renee Crichlow MD 12:38
Yeah.
Jordan Cooper 12:48
that are often underrepresented in medical research, right? Historically, medical research is white males. And so underrepresented, not only medical research, but AI training data sets. So are health centers sitting on an underappreciated asset, and what could they accomplish if they could securely combine and analyze their data across organizations
And then what's preventing that from happening?
Renee Crichlow MD 13:13
Well, a lot of things are preventing that from happening. What could they benefit from and how could they do it? It would really have to be about building trust with patients, because one of the challenges we have with caring for low-resourced communities that have been marginalized and basically
Jordan Cooper 13:31
Mm.
Renee Crichlow MD 13:32
structurally oppressed for generations is trust. And the idea that when we say it, it was like, hey, it'd be great if we could get all this information combined and we can evaluate people at scale. It could really contribute to improving both the AI infrastructure
and our clinical infrastructure, that would be great. What our patients here is, you're using me as a guinea pig.
Jordan Cooper 13:58
So is this like going back to the Tuskegee Institute and Johns Hopkins syphilis things in the mid 20th century?
Renee Crichlow MD 14:04
Yeah, it's a very good example of it. And this is the thing that a lot of people don't understand is there's a way out of that. And it's about investing in those communities. It's about saying, okay, if we're going to extract this data, what are we giving back to these communities? If you're looking to extract that data, great.
Jordan Cooper 14:15
Mhm.
Renee Crichlow MD 14:25
How many how many scholarships for, you know, electrical engineering are you putting into these neighborhoods? You know, what are what is what is your investment going to be if you're going to extract? What they've seen too often is just a one-way St. I'll give an example. I was on a vaccine committee.
for equity during the pandemic for a very large corporation. And they're like, how can we get more minorities involved in the clinical trials? And I said, you need to go to historically black colleges and endow a bunch of science positions.
Jordan Cooper 14:48
Mhm.
Renee Crichlow MD 15:02
because they will see that you're investing in the community, not just using them for their data and their body and their information. So it's not a one-way St. and trust has to be built from the investment, not the extraction.
Jordan Cooper 15:19
So speaking about being in the community, I understand that while you were CMO at Cobb and Square, you brought preventive healthcare into neighborhood barbershops. What does that tell us about the limitations of traditional healthcare delivery and what other unconventional places or partnerships might be more effective than just building another outpatient clinic?
Renee Crichlow MD 15:39
Yeah, so one of the reasons we did the barbershop, primarily blood pressure approach, was a lot of it is go where the need is. And again, demonstrating that, honestly, they don't have to trust us. We have to do trustworthy things. And one of the trustworthy things you can do is go where people are,
and provide care that people need. The barbershop approach was taken from some very good evidence that had been going on in the last decade or so. And what other things we do, things like we have a mentorship program for youth in our neighborhood that are interested in healthcare careers.
We go out and we're in schools. We're in the places where the people are, as opposed to saying, the old way is, you have to come to us if you want us to care for you. And then they're like, why aren't they coming to us? Well, you haven't demonstrated that we're really valuable. You show people value by giving them their focus and attention. That shows respect.
Jordan Cooper 16:29
Mm.
So 2 follow-up questions. One, just to clarify, I'm not sure if there's a distinction between barbershops and beauty parlors. Were the barbershop blood pressure just for men or was it also for women?
Renee Crichlow MD 16:50
So the barbershops, one of the reasons that was chosen is that men, I'm not sure you're aware of this, men are least likely to do preventive care unless their wife sends them. When I'm in a room with a patient, if he's an older gentleman, my number one question, what brings you here? The number one answer, my wife told me I had to come.
Jordan Cooper 16:59
Huh.
Yeah, alright.
Renee Crichlow MD 17:11
So yeah, the barbershops have a predominance of men that are there and a predominance of folks who don't necessarily participate in their health care on a regular basis unless there's an emergency.
Jordan Cooper 17:23
Yeah. And then the follow up thing is, I think over the, since the Affordable Care Act passed in 2010, the, yeah, the, yeah, there was a longer name for it. I forget. Patient Protection and Affordable Care Act, I think is its name. When it passed, you began seeing more investment in value-based care.
and shared savings programs and alternative payment models. Would you say that that legislation indirectly led to more health systems seeking to go provide care in the community where patients are? Or is there some other motivation for why health systems
aren't just doing, as you referred to, the old way of saying, here we are. Is it because health systems are now financially, have something financially to gain from maintaining population health as opposed to the previous fee for service model where just we get paid for more volume and we don't care if people who don't come in what their health outcomes are?
Renee Crichlow MD 18:23
Yeah, no, I think that the ACOs, the Accountable Care Organizations that have arisen from the ACA, have done, it's a sea change in the approach of larger health systems when it comes to preventative care, when it comes to using the opportunity to treat people with their chronic illnesses to not just
treat that chronic illness, but also do their preventative work and do their continuity care. It has been very important. In this country, health follows dollars.
Jordan Cooper 18:54
Mhm.
Renee Crichlow MD 18:54
And getting those dollars in a place that says preventatives, work, screening, vaccination, engaging with chronic illness on a regular basis, when they put dollars around that, oh, guess what? Health systems found it more interesting and started engaging with it more. It's not just about pulling people in so that you can do procedures on them.
It's about you can actually gain funds by doing more broader care.
Jordan Cooper 19:19
So just kind of, I guess, a topic on payment, right? So imagine you could redesign, you know, incentive, well, it's always about looking at the incentive. So imagine you could redesign how primary care is financed from scratch. I guess the real question is about adverse or moral hazard and
What would you stop paying physicians to do? What would you start rewarding? What surprising consequences might follow? Basically, I'm asking, what are people being paid to do now that isn't helpful? And what should they be doing that they're not being paid to do that an incentive would bring about positive change?
Renee Crichlow MD 19:57
Okay, great question, Jordan. The answer that I will give is primary care payment reform.
Fee for service is great if you have procedures. Excellent. You're going to do a shoulder replacement. Replace the shoulder. Great.
A person, you know, I look at the patients I take care on a regular basis. They'll come in one appointment. They have diabetes, high blood pressure, and COPD, and a family history of cancer.
Jordan Cooper 20:24
Mm-hmm.
Renee Crichlow MD 20:25
This patient needs to be in my office and I need to see them and I need to be seeing them on a regular basis. Let's take the other patient. I have hypertension. I'm 30 years old. I wear a wearable. I wear 2 wearables. I have people that wear 2 wearables.
And, and it's well controlled on one medication.
As a primary care physician, I don't get paid for that patient unless they come into the building.
Jordan Cooper 20:49
Yeah.
Renee Crichlow MD 20:50
And I have a significant number of my patients that are that kind of healthy. It's well-controlled chronic illness. With prospective payments as opposed to fee-for-service payments, I would be able to have a team of my staff engage with that patient on a regular basis to keep them out of the office.
Jordan Cooper 20:54
Mhm.
Mhm.
Mhm.
Renee Crichlow MD 21:10
but keep them healthy.
Prospective payments with a per member per month has allowed, one of the things they did in Massachusetts with the 1150 Medicare waiver was do some prospective payments through Medicaid. And that Medicaid waiver has had allowed us at Codman to bring in a large amount of team-based care.
Jordan Cooper 21:16
Mhm.
Renee Crichlow MD 21:32
So that my nurses would be calling that person with the hypertension saying, how are the medications going, any side effects, what, and their wearables downloading into our system on a regular basis so that we can see if there's any change or creep. And if it's getting better and they're well controlled, don't bring that person into the office.
Jordan Cooper 21:44
Mhm.
Renee Crichlow MD 21:53
Call them up, and if they need blood tests, put in the blood test. Let them come in. We'll talk about the blood test. If they're normal, they don't have to come in. That makes room in that office for that person that has the out of control diabetes, the out of control hypertension, the out of control COPD. And it allows me to spend more time with that patient so I can get that patient under control.
So prospective payments allow care to be done in the place that the patient needs it to be done, and it allows the clinics to stay open.
Jordan Cooper 22:23
So, Renee, I'd like to, we're approaching the end of the podcast episode. And, you know, you've practiced medicine in rural Montana, urban Minneapolis, Boston. You've worked to change healthcare from inside the exam room through policy. You're working with AI, you're working with different payment models.
I'm wondering kind of what is something that you've, what are your lessons learned? What are some, what is something that you would like to kind of a new direction you'd like to take the healthcare delivery system or a lesson you'd like to share with listeners who may not have the breadth of experience that you bring to the table?
Renee Crichlow MD 23:05
Well, I'll tell you what I tell the medical students right now. You've entered medicine at the most amazing time possible. We are on the brink of redesigning the way that we deliver care. And we are hitting that because of the ability to make some payment changes through payment reform.
Jordan Cooper 23:09
Yeah.
Renee Crichlow MD 23:24
but also through the medications that are currently here. We're in a situation with the GLP-1s and their other derivatives are going to change the face of what hospitals are going to be seeing. When I'm on the hospital service and I'm admitting patients, it's folks with out of control CHF, it's folks with diabetes causing them to have foot ulcers.
needing care, those kinds of patients are going to be fewer and fewer.
We're in a situation where we have the ability, when I talk about primary care, we do health, we do prevention, we do injury and illness. The health is how you are now. The wellness is prevention, screening, vaccination, well child, these kinds of things.
The injury is acute or chronic injury, and the illness is acute or chronic illness. If we have the ability to take care of that in the way that we know we can in primary care outpatient, that makes room in the hospitals for the people who actually need to be in the hospital. The folks who are doing the shoulder replacements, the folks that are doing the knee replacements.
Jordan Cooper 24:11
Who?
Mhm.
Renee Crichlow MD 24:30
We can, we have, we're on the verge of being able to keep people with chronic illnesses healthier longer. That's going to change the face of the way medicine looks. It's going to be brought in by payment changes and the medication changes.
Jordan Cooper 24:37
Mhm.
Renee Crichlow MD 25:00
Thank you, Jordan. Anytime.
Jordan Cooper stopped transcription
