Will ICUs become hospitalists' turf?

Expert considers new intensive care staffing models.


Quantity is not the same as quality, at least when it comes to intensivists, according to a recent study.

But that's actually good news for U.S. hospitals, given their limited supply of critical care specialists, according to study author Jeremy Kahn, MD. His research, published in the November 2015 Critical Care Medicine, found that ICUs with high-intensity daytime intensivist staffing did not have significantly lower mortality rates than lower-staffed ICUs.

Instead, other factors, such as interprofessional rounds and care protocols, may be more crucial to keeping ICU patients alive, according to Dr. Kahn, who is an associate professor of critical care, medicine, and health policy at the University of Pittsburgh.

He recently shared his thoughts on current and future ICU staffing and practices with ACP Hospitalist.

Q: What led you to study the impact of intensivist staffing?

A: Patients admitted to the hospital are increasingly sick, and the overall incidence of critical illness is rising. As a result, hospitals themselves are looking more like ICUs. This is creating a tension where we have more and more critically ill patients in the hospitals but we struggle to provide effective and efficient care for them. The gold standard for ICU staffing is intensivist physicians, but there simply aren't enough intensivists to go around. We can't put a trained intensivist in every single intensive care unit. It's simply not a feasible policy direction. My hypothesis in this research is that rather than re-demonstrating the value of intensivists, we need to acknowledge that there aren't enough intensivists to go around, and figure out the mechanism of benefit of intensivists and how we can extend that benefit to other hospitalized patients.

Q: What did your study figure out?

A: The study shows that in the setting of interprofessional care and in the setting of advanced protocols, that the actual mortality benefit from intensivists is small. This is not to diminish the role of intensivists, but to simply say there are other ways to achieve high-quality care . . . like care protocols and interprofessional rounds. We might be able to implement those things more easily than expanding the role of intensivists, giving higher-quality care to critically ill patients without having an intensivist in every ICU. The hospitalist model might very well work under this hypothesis.

Q: What would be the role of hospitalists?

A: You could have an intensivist ICU medical director there to coordinate interprofessional rounds and implement and evaluate protocols, but daily care provided by hospitalists. In this scenario, intensivists could be there to see the most sick cases. This model is analogous to a generalist-subspecialist model that has worked when we define specialists by disease. Take hypertension. We acknowledge that for the less sick patients, generalists are quite adept at caring for those patients, and it's not feasible for every patient with hypertension or heart disease to see a cardiologist. So we reserve cardiologists for the sickest patients and we also use them to set protocols and policies regarding the care of cardiovascular disease.

You can think of intensivists as subspecialists of hospital medicine. Both because [hospitalists] are quite skilled in the care of hospitalized patients and because it's not feasible to have an intensivist everywhere, hospitalists would responsible for most day-to-day management of patients in the ICU and we would reserve intensivists for care of the sickest patients and for setting overall ICU policies and procedures. It's actually not that innovative an idea—it's just rethinking the generalist/subspecialist relationship for hospitalized medical patients in a way that we've done already for outpatients.

Q: Would that model require redistribution of the existing intensivists?

A: Possibly. Right now they're mostly in urban academic hospitals, and we have to think about redistributing them, so creating incentives so some more will work in rural areas. You'd also have to empower hospitalists to expand their role in intensive care, perhaps even in urban academic hospitals. The role for hospitalists would grow in some hospitals, and the role of intensivists would grow in other hospitals, but together we'd be right-sizing critical care.

Q: What would be the first step in shifting to this model?

A: This notion that we might achieve similar outcomes for critically ill patients by staffing the ICU primarily with hospitalists and nonphysician providers, overseen by intensivists, is a hypothesis that needs to be tested. It seems feasible and practical, and the data thus far would support it. It appears that at least part of the benefit of intensivists in the ICU is due to not who they are, but what they bring. They bring interprofessional rounds, they bring protocols and care pathways. However, we need more research proving that an intensivist-directed, hospitalist-led model of critical care is safe and effective.

Q: How does telemedicine fit into this?

A: Very unclear thus far. Telemedicine is a technology in evolution. I think there is compelling evidence that it can also improve the quality of care in small rural hospitals, but just not consistently so.

One way it fits into the paradigm I've described is that if it appears as if we can achieve very good outcomes for critically ill patients by a hospitalist providing day-to-day care overseen by an intensivist, then it's worthwhile asking the question: Does that intensivist need to be physically present? Could the intensivist provide their services remotely, in terms of checking up on the most severely ill patients in the ICU, implementing protocols and care pathways, and facilitating interprofessional rounds? Then we can really get the benefit of the expertise of an intensivist everywhere without having to physically put an intensivist everywhere.

The early data are quite favorable in support of telemedicine. It's just that where telemedicine has been tested most thus far has also been where intensivists are at the bedside.

Q: What other interventions in this area need more attention?

A: Interprofessional rounds are a quite underutilized strategy for improving critical care. More and more hospitals have them, but we also don't understand how to maximize their quality and their role—how to get the most bang for the buck. We increasingly recognize that providers like clinical pharmacists and respiratory therapists [and] nutritionists have unique expertise that physicians don't always have. It's not efficient for an intensivist to acquire all that expertise, so I think the interprofessional model of care—although strongly supported by professional societies like the Society of Critical Care Medicine—it's just at this point not fully understood and probably not optimally implemented.

Q: Will there be a single model that's best for all ICUs?

A: Not at all. Every ICU will need to find a solution that's best for them, within the context of the broad organizational models I've described. Some ICUs will need 24-hour intensivists, some ICUs will just need daytime intensivists, and some ICUs can get by with no intensivists, but with intensivist oversight from a managerial perspective.

Q: What will determine which category a hospital fits into?

A: I think ultimately it will be case mix and the availability of other services in the hospital. There are some diseases and therapies that will always require the presence of a bedside intensivist day to day. These are things like severe acute respiratory failure, multisystem organ failure, and shock. These are the kind of patients where it is, in my opinion, essential to have the expertise of an intensivist at the bedside. Additionally, the data would support that, because we know there are volume/outcome relationships for those types of conditions, meaning that the more patients you see, the better you are.

Q: What's the biggest take-away message for hospitalists?

A: There are important roles for hospitalists within the intensive care unit. It's a mistake for hospitalists to view their jobs as ending at the ICU door, even in hospitals that have intensivists. As our research evolves, it behooves the hospitalist movement to engage intensivists to help us think through these innovative partnerships for providing physician staffing in ICUs.

Q: How do the involved physicians, hospitalists and intensivists, feel about this idea, in your experience?

A: I think it varies widely. My experience has been that most intensivists are open to alternative models of care, and at the same time, most hospitalists are eager to engage with their critically ill patients. I don't think the community is of one mind yet. I do think that what will drive these alternative and more novel models is necessity more than desire. Once we get past this idea that we can't extend the intensivist physician staffing model to every ICU in the country, then the only option is a more collaborative approach. We can both elevate the role of hospitalists and other general practitioners in the ICU while not diminishing the role of intensivists. I think it actually is win-win.