Why eliminating ‘hospital purgatories’ would save billions of dollars

BY Sarah Corapi  December 3, 2013 at 4:02 PM EDT

Making observation services in U.S. hospitals more effective and efficient could save the system billions each year, according to a new study in the journal Health Affairs. Photo by Mario Tama/Getty Images

It doesn’t take a brain surgeon to figure out what goes on in an emergency department, intensive care unit or pediatric wing. But what about a hospital’s observation services? Few Americans understand them or even know they exist. Which is why — as health care costs continue to climb by the year — it might come as a shock that these are the very services that could end up saving hospitals billions each year.

Put simply, observation services are designed to determine which patients can be safely discharged from the hospital and which should remain for a longer stay. All hospitals provide some form of observation services — typically to emergency department patients who might be too sick for outpatient treatment but not sick enough to be fully admitted to the hospital.

“I call them the ‘six- to 48-hour patients,’” said Dr. Michael Ross, associate professor of emergency medicine at the Emory University School of Medicine. “These are patients who need more than six hours of emergency care but who might only need to spend 24 or 48 hours in a hospital if actively cared for. But hospitals are so focused on inpatient care or outpatient care, and these patients really fall between the cracks.”

Ross, lead author of a new study in the journal Health Affairs, maintains that inefficient and disorganized observation services are putting huge drains on the country’s already taxed emergency health care system. Between 1997 and 2007, visits to emergency departments in the U.S. grew at double the rate of population growth, while the number of emergency departments decreased. The heightened demand and shrinking resources didn’t bode well.

And observation services that are administered without a dedicated setting or defined care guidelines are only adding weight to the load. It’s estimated that two-thirds of hospitals in the U.S. use inefficient practices, which translates to longer stays for patients and an additional $331 per patient in costs for the hospitals themselves. It also translates to poorer care conditions for some, who may find themselves in “observation purgatories” awaiting better treatment.

By focusing on the way they deliver observation services, the Health Affairs article concludes, hospitals could see big savings. Implementing designated observation units with clear care protocols could get people back to their normal lives more quickly and keep them out of the hospital for longer. They could provide up to 28 percent shorter lengths-of-stay and potentially a 44 percent lower probability of subsequent inpatient admission.

In total, it could all add up to nearly $1 billion per year in savings for hospitals — or much more. Some patients who would normally be admitted to inpatient care could be eligible for observation care, if it’s provided correctly, and that could mean a possible savings of $5.5-$8.5 billion annually.

But does changing the way observation services are provided really make that big of a difference? Are protocol-driven observation units feasible in hospitals throughout the nation? And worse — could they even prevent patients from getting the inpatient care they might need? PBS NewsHour recently spoke with Ross about the potential for improving observation units nationwide and its money-saving potential.


PBS NEWSHOUR: Dr. Ross, thank you for joining us. Let’s start with some context. What is the role of observation services in a hospital setting?

DR. MICHAEL ROSS, Emory University School of Medicine: Observation services are essentially outpatient services, provided usually to emergency department patients to determine if they need to be admitted to the hospital as an inpatient for a longer stay.

Patients admitted to emergency departments usually spend about five or six hours there, since emergency departments are not designed to take care of patients for periods of time that are much more extended than that — otherwise they become overcrowded.

After the five- or six-hour threshold, a decision has to be made to move the patient on to another setting or not. What this has done is create an orphan population of patients, for lack of a better term — I call them the “six-to-48-hour patients.” These are patients who need more than six hours of emergency department care but who might only need to spend 24 or 48 hours in a hospital if actively cared for. But hospitals are so focused on inpatient care or outpatient care, and these patients really fall between the cracks.

NEWSHOUR: What do you mean “fall between the cracks”? What happens to these patients?

ROSS: What’s occurred in hospitals across the U.S. is that these observation patients are often placed on a floor where they’re mixed in with five-day patients, and they often don’t get the most optimal care in terms of timing and efficiency. If we’re trying to determine — within 48 hours — which patients need to be admitted to the hospital for a longer stay and which patients can be discharged, their care needs to be actively managed.

Some things that hospitals can do, and have already started to do, is to either have a dedicated unit for these patients, and/or provide defined care guidelines for these patients.

The best practice is to combine the two together, which we call a “Type 1″ observation unit. About a third of hospitals in America have a dedicated observation unit, and about half of those hospitals treat patients using defined protocols — so about 1 in 6 hospitals are using this Type 1 practice.

But most observation patients in the U.S. are treated using the most ineffective practices — in a bed anywhere in the hospital, with inconsistent and inefficient care. News stories, research and Medicare claims all suggest that this setting is the least optimal for these types of patients.

NEWSHOUR: What are these sources saying, then, about the treatment patients get from disorganized observation services? How bad can it be?

ROSS: Most observation patients in the U.S. are managed in a bed anywhere in a hospital at the discretion of the treating physician. Roughly 7 percent of these patients will remain in observation for more than 48 hours, compared to less than 0.1 percent of patients managed in a Type 1 unit. Previous news stories have described problems that arise for these 7 percent patients. This includes confusion that they were never admitted as an inpatient even though they were in an inpatient bed, increased out of pocket costs, and potentially having to pay costly nursing home bills.

The two prime drivers of the growing length of stays for these patients are the payment policies and the setting. Regarding the payment policies, the criteria required for inpatient admission had become increasingly strict, forcing patients that should have been inpatients to be kept as outpatients. This policy issue prompted (Centers for Medicare and Medicaid Services) to write what has been called the “two midnight rule,” making it easier to admit patients whose observation stays would go beyond 48 hours.

However, this rule does not address a bigger issue — the setting. Our Health Affairs study shows that most of the time, savings between Type 1 units and disorganized observation settings occurs between 24 and 48 hours. Unlike an inpatient bed spread anywhere throughout a hospital, Type 1 units are designed to treat patients and make admission decisions in less than 24 hours.

NEWSHOUR: And these observation service units with defined protocols for care — Type 1 units, as you call them — can really prevent these bad scenarios from playing out?

ROSS: These “observation horror stories” are driven by excessive length of stays. Type 1 units by design have lower length of stays. Just as ICUs are designed to provide for the needs of critical patients, Type 1 units are focused on the timing and decision-making nature of observation patients.

NEWSHOUR: Your study also says that Type 1 units are the key to getting patients out of the hospital sooner and keeping them out of the hospital for longer. Why is that the case?

ROSS: When a patient comes through an observation unit, they’re often receiving testing or treatment for a critical condition, like a heart attack or stroke. And while they’re utilizing observation services for this, it can be a great opportunity for what we call a “teachable moment” — a time to educate them about how to prevent future heart attacks and strokes, or how to take better care of their health in general. So this certainly can help patients stay healthier and remain out of the hospital or emergency department for longer.

NEWSHOUR: What about the potential cost savings? Where do they come in?

ROSS: Type 1 units can provide big monetary savings, as well, and it all comes down to efficiency: the things that are essential for the treatment of a patient are identified, and then the hospital’s resources are aligned to make sure that they happen promptly — so that rather than taking three, four or five days to complete the treatment, it all happens within 15 hours.

Within an average of 15 to 18 hours, it’s identified that a patient can be discharged and sent home safely, or that they need to be admitted to the hospital. So the real savings come through that efficiency — that decrease in length-of-stay. In our study, we estimated that applying this type of care to all observation patients in the U.S. would save the U.S. health system about $950 million a year. Another study showed that patients who are cared for in the worst type of observation setting — no dedicated unit with any defined guidelines for care — causes hospitals to lose an average of $331 per patient.

NEWSHOUR: But savings can also come into play when you consider the number of patients that might be saved from spending unnecessary amounts of time in inpatient care, correct?

ROSS: Sure. These savings can also apply to patients that have already been admitted to hospitals as inpatients, but who have conditions that really could be treated in the observation units. These patients make up about 11 percent of all hospital admissions per year in the U.S. If these patients were given shorter lengths of stay in Type 1 observation units to determine if they really needed more hospital care, it could save the U.S. health system about $5.5 to $8.5 billion per year.

NEWSHOUR: It sounds like these protocol-driven observation units have a lot of advantages. But are there any drawbacks to implementing more of these in hospitals across the U.S.? Could they hold people back from getting the real inpatient care they might need?

ROSS: No, not at all. In fact, it could lead to getting better care. For example, some Medicare patients who are admitted to the hospital may need to go on to a nursing home after they’re discharged. But they need to spend at least three days as an inpatient in the hospital in order to get their first 21 days in nursing home care covered. What’s happened in some cases is patients waste a lot of time in inefficient observation before they’re admitted, and then because of that, they end up spending less than two days as an inpatient. When they go to the nursing home, the benefit isn’t covered.

The benefit that guideline-driven observation units offer in this situation is if the decision to admit is made within 15 hours instead of 4 days, those patients could get into inpatient care faster and are less likely to lose their nursing home benefit. So does it hold people back from getting the inpatient care they might need? No. Type 1 units are driven to treat and diagnose patients efficiently and to quickly decide if they can be admitted or discharged.

There have even been three studies where patients were randomly placed in these Type 1 units versus a regular hospital bed, and in all three studies, patients were more satisfied with the care they received in this setting. Patients like feeling that their care is being actively managed, not passively managed.

NEWSHOUR: But what about cost concerns for patients? It might save hospitals money, but aren’t some patients — Medicare beneficiaries especially — going to have higher out-of-pocket costs?

ROSS: There’s definitely a concern among Medicare patients that if they’re placed in observation, their out-of-pocket costs will be more than if they’re admitted as an inpatient. This concern led to two class action lawsuits against Medicare. And as this has played out, the Office of the Inspector General has investigated observation services to see if this is true.

What they found is that actually, for 94 percent of patients, their out-of-pocket expenses are less than if they had been admitted as an inpatient. The 6 percent that had higher out-of-pocket costs were overwhelmingly those who had coronary stents or angiograms — both things that are really not done in Type 1 units as a rule. So there’s been an exaggerated belief that observation is more expensive, when, in fact, for 94 percent of Medicare patients, it would be more expensive to be admitted as an inpatient.

NEWSHOUR: What would it take to have these kind of units widely implemented in hospitals across the country? Is it even feasible?

ROSS: It’s absolutely feasible. It’s estimated that between 10 to 25 percent of patients that stay in the hospital following an emergency department visit fall into this category. Some hospitals have wondered if they have enough patients for a dedicated unit. I can’t answer for all hospitals, but I think it’s likely that most hospitals have enough of these types of patients to support a Type 1 unit.

Think about it this way, we would never think of placing an ICU patient in a bed on a regular hospital floor — it’s a service that requires a distinct and separate setting. Observation services, on the other hand, are the only evaluation and management services recognized by the American Medical Association that are not consistently provided in their own dedicated settings. There’s just been kind of a learning curve for hospitals to realize that these really are distinct patients who really need their own distinct setting.

And what would it take to make more of these dedicated observation units a reality? I think the best, first step would be for Medicare to simply ask hospitals to report the settings in which observation services are provided — if it’s a Type 1 unit or not. Then they could look at the outcomes, the length of stay, and the costs associated with different levels of care. What they’d likely find is that Type 1 units offer the best outcomes, and from there they could create payment incentives for hospitals to adopt these types of units. This would be the strongest driver of change.

NEWSHOUR: Michael Ross, thank you very much for joining us.

ROSS: Thank you.