Zuckerberg San Francisco Basic Hospital (ZSFG) has achieved a big turnaround in readmission charges by combining a predictive threat mannequin in Epic, standardized care pathways, and proactive outreach to at-risk sufferers. Lucas Zier, M.D., M.S., director of cardiovascular high quality and outcomes at ZSFG, spoke with Healthcare Innovation about this initiative in addition to efforts to share AI improvements with different security internet well being methods.
Healthcare Innovation: May you speak a bit of bit about why ZSFG and different security internet hospitals have struggled with readmission charges previously and and a few of the monetary repercussions that has?
Boiled: Popping out of the Reasonably priced Care Act, one pay-for-performance metric entails readmissions, and the hospital readmission discount program was instituted. I feel it was very effectively intentioned, however the impact was that well being methods that are likely to serve susceptible and underserved sufferers, by definition, look after sufferers with medical challenges, but in addition a variety of opposed social wants. That tends to result in a affected person inhabitants which is extra vulnerable to readmission. The downturn impact of that’s that well being methods that serve our most susceptible sufferers are likely to get penalized most severely for elevated readmission charges.
San Francisco Basic Hospital, as of 2016, had a few of the worst readmission charges within the state of California when in comparison with different security internet hospitals. The repercussions of that had been several-fold. One was that it was imperiling about $1.2 million in funding that we had been utilizing for medical care applications. Additionally, we realized that elevated readmission was reflective of some sub-optimal outcomes that we had been having, significantly in sufferers with coronary heart failure, which was was our greatest driver of readmission. We had elevated mortality charges in comparison with different security internet well being methods, and we additionally had fairness gaps in care. Particularly, the Black/African-American coronary heart failure inhabitants had worse outcomes in comparison with our basic inhabitants.
HCI: Did you develop or customise a predictive mannequin inside Epic?
Boiled: Sure, we took a multi-pronged strategy. We localized Epic’s readmission mannequin to our well being system. However we acknowledged that simply offering an end-user with a threat prediction was not going to be adequate to enhance well being outcomes. We took that predictive output and developed a choice help ecosystem. We mixed that with know-how that Epic supplies, the place you’ll be able to basically hyperlink collectively these logic-based blocks inside Epic. We recreated the guts failure tips and we had been capable of floor resolution help to suppliers on the level of care that was personalised to sufferers.
We might make suggestions about medical care and drugs, and we deal with each medical wants and a restricted slice of social wants — significantly substance use, was a giant problem in our coronary heart failure inhabitants. One instance of resolution help could be if a affected person screens optimistic for for methamphetamine use, then we had been capable of set off a referral to our habit care group. The predictive mannequin allowed us to risk-stratify who’re our highest threat sufferers, so we might floor that data to suppliers after which immediate them to position prioritized referrals for particular person sufferers.
We had a inhabitants well being administration group for coronary heart failure, however they did not all the time know who to give attention to. The advantage of the predictive mannequin was that it ran throughout our whole coronary heart failure inhabitants, so our group might take a look at this dashboard that we developed, and see who’s anticipated to be at excessive threat for readmission. That group might proactively look after these sufferers locally to attempt to stop a readmission, earlier than it occurred, versus reacting to a readmission that already occurred.
HCI: Do a sure share of those sufferers not have a main care relationship? And is that a problem, so far as following them locally?
Boiled: That could be a actually good query. Sure, some sufferers haven’t got a main care relationship, and that is among the potential cease indicators earlier than you discharge sufferers is to verify they’ve a referral to main care, as a result of we all know sufferers who get referral to and are seen by main care are likely to do higher once they depart the hospital. However one of many advantages of the predictive mannequin is that if a affected person is anticipated to have a excessive readmission threat, then when a supplier locations a follow-up referral for cardiology, that referral will get prioritized inside our referral queue.
HCI: Was there a timeframe over which you studied the influence of this intervention?
Boiled: Principally, we had recognized the issue way back to 2015. At that time, we had been utilizing LEAN methodology to begin to develop some countermeasures. These preliminary pilots had been efficient, however they weren’t scalable. We went by way of an Epic implementation in 2019 and that is after I began to appreciate that perhaps there’s a possibility to make use of the EHR for large-scale, system-wide efficiency enchancment. Our outcomes are actually from 2019 to 2024. It wasn’t a randomized analysis, however we did what we name an interrupted time-series evaluation. Basically we checked out numerous charges of issues earlier than we applied the device, after which after we applied the device. The particular issues that we checked out had been our readmission charges and our mortality charges amongst our coronary heart failure sufferers. We additionally regarded particularly at outcomes inside our Black/African American inhabitants, as a result of we knew we had important fairness gaps in these outcomes.
General, there was a 6% discount in mortality. We additionally in contrast our outcomes to 5 different peer security internet hospitals in California, and we discovered that we had, in comparison with these different well being methods, a big discount in mortality amongst our coronary heart failure sufferers. You possibly can by no means be 100% positive until you will have a randomized trial, however we really feel that we will confidently conclude that it was the impact of the device.
We additionally in contrast our readmission charges to different security internet hospitals throughout the identical time interval. And for a lot of of these hospitals, their readmission charges went up, whereas ours went down. At our peak, our readmission charge was about 34% and at its nadir it was about 19 p.c. So we had an over 13% discount in readmission charges, taking us from mainly the worst, relying on the month that you simply take a look at, to top-of-the-line amongst security internet hospitals. And basically, we have inverted this pattern with our outcomes amongst Black/African-American sufferers and our basic coronary heart failure sufferers. We had important inequities, significantly in readmissions, and by 2022, we had utterly inverted that pattern. There was no distinction in readmission charge.
It has been effectively documented that generally when well being methods attempt to institute readmission discount applications, that comes on the expense of different outcomes. It is not unusual for well being methods, for instance, to see reductions in readmission with a slight enhance in mortality, and we had been actually targeted on avoiding that. That was why mortality was an essential metric for us. We had been capable of scale back readmissions whereas additionally lowering mortality. In our minds, it is not an efficient final result should you obtain a well being system metric, however you place your sufferers in danger.
HCI: Let’s return to what you talked about earlier — this consortial work with the opposite security internet hospitals. Is that to unfold this specific intervention at different security internet hospitals or is it broader than that?
Boiled: The overarching thought is that we consider that this sort of know-how has important alternative to enhance well being outcomes in well being methods that aren’t significantly well-resourced.
This value about $1 million to develop, however as a result of we’re capable of hit readmission metrics, we saved near $8 million.
One strategy to this readmission discount program is simply rent 10 extra folks and ship them out locally, however security internet well being methods cannot do this. We predict there’s an actual position for AI and machine studying to handle challenges inside these well being methods. However $1 million will not be cheap. We’re fortunate at San Francisco Basic that we have now a basis that helps to help this work. And we have now a relationship with College of California, San Francisco, which creates some infrastructure to assist us to construct a few of these custom-made instruments. However most security internet well being methods don’t have that sort of help. So the collaboratory was based with the intent of taking a few of the well being methods, like Parkland which has an analogous setup to San Francisco, and Grady does as effectively, the place we will function an incubator for the event of those applied sciences.
If we will develop them, we must always attempt to disseminate them and share them with different well being methods. It is nonetheless very a lot in its inception. What we’re specializing in proper now’s a generative AI device that just about capabilities as a digital social employee to attempt to deal with affected person social and behavioral wants inside these populations. We hope that we will disseminate learnings. We will additionally disseminate know-how, and we will make it obtainable to one another at decrease value.
