Jenifer Leaf Jaeger, MD, MPH
As healthcare systems continue to shift from fee-for-service to value-based care models, it is important for providers to meet the patient where they are to improve their overall health and well-being. Addressing social determinants of health, including homelessness, lack of food, and transportation, can be crucial, especially for patients who have a chronic condition.
Jenifer Leaf Jaeger, MD, MPH, senior medical director at HealthEC, spoke with HCPLive
® about addressing population health and how it could improve outcomes and quality of life for patients with chronic conditions, and the importance of data, technology, and community partnerships.
Editor’s note: This interview has been lightly edited for style, length, and clarity.
HCPLive: How can health systems leverage technology to improve population health?
That's the only way to do it. If I have 1 patient record, then I have 1 patient record, and I know what happens to Bill Smith, but that's about it. If I'm able to have 1000 patient records from everybody who lives in this area, then I'm able to start taking a look at the population.
Here's a good example: Healthcare is: Bill Smith smokes a cigarette or is a smoker and he comes in with cancer or COPD
. He goes to his doctor and his doctor takes care of him. That's healthcare. Public health is saying that we know smoking increases the risk for lung cancer. People have done all the work and now we need to put out these public service announcements and messaging to our communities and try to put pressure on cigarette companies not to advertise to kids by saying smoking causes cancer or increases your risk for cancer. Don't smoke, right? That's good; that’s public health. Population health is taking a look at the people who smoke or all of your lung cancers, right? And you say, “Well, I see that they're predominantly smokers, but I also see that African Americans are much more likely to develop lung cancer or have a fatal course with their lung cancer than white Americans.
So, what is the difference? Why is there a difference by race and ethnicity? Why is there a difference in individuals who have Medicaid as opposed to private insurance? Why do I see that individuals who live in the South do worse than the individuals who live in the North?
Everybody knows not to smoke cigarettes, but for the folks who smoke cigarettes, you know who is getting sicker. Is it that the messaging works better for 1 section of the population and so they've quit smoking, whereas the messaging didn't get to the other population? Well, that's another part of the population. So, what about that messaging? Is it that it's targeting 1 particular culture or group of people and it's not being culturally or linguistically sensitive to address the other population? Or is it that doctors respond to complaints from certain patients faster and more aggressively than with other kinds of patients?
You are able to look at disparities between subcomponents of your population and say, “What's the difference?” so that you can actually have an impact? And if you say that if somebody has housing insecurity and is living in a shelter they are for example, three-times as likely to present in a later stage than individuals who are smokers, but have their own car, then I can say I know that transportation and housing are an issue, and so, I've got a smoker here and these are the social determinant assessments that I need to perform with these folks. Then I can give them the social services that they need and have a care manager who links them directly to the services, and then follows up to make sure that they're able to make their appointments and have culturally sensitive, anti-smoking, support services.
Now you have the impact of improved care at a patient level and improved outcomes at a population level because you're understanding where the risk factors come from. But you only know that if you're looking at 10,000 records—that's the only way to actually have valid data. It takes stepping away from the individual patient and looking at it from a 30,000-foot level, getting very granular in your data analysis, and then taking what you learn from that analysis and then applying it to the impact of the individual patient. That's the only way to do population health, and that requires leveraging technology in a very sophisticated manner—being able to pull in EHRs from all the different hospitals where your patients so you are able to aggregate all the data and do something with them.
Population health doesn't happen without technology.
HCPLive: Is there a specific healthcare facility (ie an urgent care facility, primary care, psychiatric hospital, etc.) that might be best-suited to use technology for population health?
The pat answer is no. Each environment can pull from it what is useful for them. But I was thinking about it as you were describing it and would say that in the area of care management—not in the emergency room, not in urgent care—but working with chronic disease conditions. I think it has some of its greatest impact. I would say outside of healthcare, in terms of the drivers of this work—the payers, private, commercial insurers—the accountability that physicians and organizations have to the payers is also incredibly useful and the ability to look at data at a large scale is something that
the payers have an opportunity to manage the whole population. I think it's useful there too.
That being said, in a clinic, urgent care, or an emergency room, the focus isn’t on management long-term. That's where primary care and emergency care
takes priority over public health or even population health in. If someone comes in and they’re not breathing, you need to get them breathing and that's urgent.
I think it is useful in terms of educating the providers. I've worked in urgent care and emergency rooms in vulnerable, underserved areas, and certain populations are at greater risk for showing up with an asthma attack or COPD. These folks are often living in publicly subsidized housing or shelters and their access to care is limited and sometimes they have to take 2 or 3 buses just to get to an urgent care. You have to be mindful of the triggers that set that asthma attack off in terms of giving your anticipatory guidance and patient instructions after they leave. But that I think is not so much doing population health, so much as utilizing the data that we've gleaned from the work the population health does, and then saying, Oh, this is what we've learned from all this work, and I can now implement this in my patient care and my management when they were in and out of my emergency room.”
So, chronic care management is probably where we see it best, in large hospitals and healthcare systems.
HCPLive: How can a focus on population health improve outcomes for patients with conditions like asthma or diabetes?
For providers and payers, population health is an investment and it becomes a useful and sustainable investment because value-based care and contracting. If you can show the utility in terms of improved quality of care, improved outcomes, and decreased cost, your system is paying for itself.
One of the CMS quality measures with regard to diabetes
is taking a look at hemoglobin A1C. Poor control is marked by a high A1C level—the cutoff is 9. If you are able to take a look at all of your individuals that come to your hospital system and you get the lab results from every single patient who comes in and you have the their ICD-10
codes, you can sort for all your diabetics and then within that you can sort for all of the hemoglobin A1Cs that had been done in the last 3 months and take a look at who has an elevated level above 9. And then you have a care coordinator who is able to access those records and then reach out to the individuals and say, “Hey, we need to bring you in for a visit.” Then they are not lost to follow-up.
Somebody who's in poor control might be that way for any number of reasons. And the first thing you need to do is identify them. Physicians who are seeing patients maybe have 10 or 15 minutes to see a patient and can't be doing this as well as seeing all the patients that they are expected to see in a day. And nurses are supporting being able to see 10-20 patients in a day. So, the care coordinators are able to go into their system at the organization, practice, or even at the particular provider level, and pull up these at-risk patients—and you can do it by any diagnosis. So, you can sort for asthma
, diabetes, COPD, and 28 different chronic conditions that have been called out by CMS.
Specifically, with our platform, we can sort for all of them and anyone that you wanted, all you need to do is identify which ICD-10
codes you'd want to look at and then you can sort for this and that. You can have this auto-populate your dashboard for care coordinators, who will log in in the morning and have for them a list of individuals with diabetes who have a recent A1Cs that are out of control. You could then even sort for the ones that have had 2 in a row that are out of control and say, “Okay, these are the 40 most critical patients that I need to address right now, and then call each of them.” You could actually make a telephone call to each 1 of these individuals and identify what their barriers to care are. “Well, I lost my insurance,” or “I couldn't get there,” or “I'm taking care of my elderly mom and I couldn't make that appointment.” Then you're able to actually do that kind of care coordination and close the loop by these folks. The individual patients are then referred and brought into clinic or immediately linked to the agencies that are in their area to support their particular needs.
You can do that with any chronic condition if you have all the data.
HCPLive: What population health trends are you seeing nationally?
The CMS 1115 waiver that allows you to address social issues, particularly housing, transportation, and food insecurities. The push may be from the payer standpoint to reduce cost. Because what we spend in this country on healthcare is not sustainable and we've been saying that for a decade and it's even worse and not sustainable. It is necessary to reduce what we spend on healthcare as a nation, and people should not have to decide whether they want to have their asthma controlled or to feed their child—they just shouldn't have to make that decision.
So, the fact that payers, CMS and various states and private insurers can now reimburse the addressing of social determinants is huge. And I hope that other private insurers follow the lead from Humana. It makes sense because it's not immediate. If you're choking on an olive pit and I do the Heimlich, I save your life immediately. That's great. And you can say, “Oh, well, we should reimburse you for that.” It's really hard to say, “Well, I'm going to get you housing and that's going to decrease your cost of healthcare in the long run and it's going to improve your health outcomes in the long run, and you're going to live longer and have a greater quality of life.” That I can say, and I can even show you data, but it's really hard to get people to put money behind that.
I think that is that is the trend that we are seeing with greater and greater states being able to step up and support that, and now with private insurers beginning to reimburse physicians for that work. What is so different about the reimbursement strategies now is that you can share in the savings of lowering costs below a certain benchmark or threshold that's set by using CMS. But you don't get to share in any of that savings if your quality isn't above a certain level. And CMS also sets those benchmarks. So, saving money doesn't do you any good. I mean, it may save you a little bit of money, but you really don't benefit from that if your quality is not also maintained at a pretty high level.
These alternative payment models, as they are evolving and getting better and better at incentivizing good care, is a trend in the right direction and I would love to see every state involved in this.
HCPLive: How can partnerships be leveraged to launch population health initiatives?
At 1 level, it's just your community agencies. One of the things that we have in Health EC’s Care Connect Pro program is assessments on the social determinants. And it's automated so that if you're answering “yes” to certain questions, it will pop up the agencies, organizations, or nonprofits, that are in the patient’s neighborhood based on their zip code and in that general vicinity. Then you can actually click and hit a referral automatically. We have various organizations pre-populated in our platforms.
So, having clinics, urgent cares, and hospitals make these connections with the agencies that are in their area is the first level of partnership.
I’m sure the agencies are happy to take any referral that you have, but if you know these groups— especially the smaller ones that are either culturally or linguistically sensitive to the population that you serve—then you can speak to your patient and say, “Okay, you know what? I'm going to send you to so and so and I want you to talk to so and so,” and you have a name and you can make the referral electronically. Ultimately, you're speaking to the patient in front of you and telling them that you’ve worked with this group and they’re really terrific and here is a referral. And that really helps.
Identifying which community resources are in your area is first. Second is being able to, at an organization level, have relationships with lab companies so that you have access to the labs for a patient that you are responsible for. So, a patient that goes to Hospital X and that we are serving, if we are able to then have access to that person's labs, no matter where they go. When they're on vacation or they were rushed to the emergency room that isn't their standard hospital, having those data is critical. That's at a totally different business use agreement level where you have hospitals and lab companies. And from our standpoint, health IT companies since that kind of relationship is also critical.
HCPLive: What challenges or obstacles might a facility face when implementing these initiatives or using these technologies to address population health?
The first thing that comes to mind is, as much as I talk about 1 big happy family, so many organizations and arenas are siloed, which makes developing the partnerships really critical. A champion or an early user or adopter is going to have to create relationships and break down silos to start.
The second is that it’s an investment in time, effort, money, and resources to implement a new kind of health IT endeavor. Even adopting an EHR can take months and months in terms of implementation. In terms of addressing population health as a whole, any implementation will take early adopters and champions that really understand it—how to use it, what utility is, and not just what buttons to push, but why would you push those buttons in a certain time.