As CMS makes another policy change, policy makers distinguish between different forms of care
Experts urge hospitals to consider the creation of dedicated observation units
As the Centers for Medicare and Medicaid Services (CMS) make yet another change in the regulations regarding "observation status," the confusion continues over when a patient qualifies or doesn't qualify for this type of care, and really what is at stake in these decisions for patients and providers. While emergency providers seek clarity on all these points, proponents of observation care lament the fact that somewhat lost in the discussion is what they see as growing evidence that when used effectively, protocol-driven observation care that is delivered in a dedicated observation unit can be a superb tool for expediting care and reducing hospital admissions. Though the regulatory infrastructure is not yet in place to require hospitals and EDs to deliver this type of focused, accelerated care, proponents make the case that there is nothing to stop providers from taking steps to bring their use of observation in line with the kind of evidence-based care that is associated with dedicated observation units.
Recognize the needs of observation patients
"The message that has to be gotten out there is that there will always be patients who need more than six hours of care and less than 24 hours of care," explains Michael Ross, MD, FACEP, medical director of Emory University Hospital's clinical decision unit, an observation unit that is adjacent to the Atlanta, GA-based hospital's ED. "We can either ignore them and push them all into EDs, leading to overcrowded EDs, or push them into the hospital, which is going to lead to overpayment for those patients, or we can say this is a very real group of patients with very real service needs and provide them with the care that they need."
What is plain to see, according to Ross, is that there is a "widening chasm" between care that is delivered in a protocol-driven ED observation unit and a bed that is somewhere else in the hospital. In fact, Ross has written extensively about what he views as state-of-the-art observation care.1However, currently there is no regulatory requirement that observation care be delivered in an observation unit or any other specific setting, so patients are often placed in beds on inpatient floors even though the patients themselves may not even realize that they are still technically outpatients.
Critics charge that the problem with this arrangement is that care tends to be less focused and that rounding may only occur once a day. Indeed, a recent study on observation care in the Archives of Internal Medicine noted that observation patients are often subjected to lengths of stay (LOS) in excess of 24 hours, and that hospitals frequently lose money on these patients. The authors questioned what role observation should even play in health reform.2
However, Ross suggests that outcomes are quite different when patients are cared for in dedicated observation units, where rounding is constant and care is driven by protocol. "This is a unique patient population, it is a recognized patient population, but for some reason, two-thirds of hospitals haven't been able to recognize the need to provide these patients with a unique setting," explains Ross. "There is a lack of understanding about what observation units are, so I think there has been a clinical need for education."
Ross suggests that hospital administrators need to consider that observation patients make up the lion's share of both misdiagnoses and malpractice lawsuits in emergency medicine. These are also the patients that cause ED overcrowding because they don't meet inpatient criteria, but providers also can't send them home, he observes. "This is the group that everybody would like to ignore, but that is the worst thing we could possibly do," he says.
Be prepared for RAC scrutiny
Another problem plaguing observation care is that patients have sometimes remained on inpatient floors for as long as five days, resulting in exorbitant fees for patients, because as outpatients, they are responsible for 20% of each individual charge under Medicare Part B. The new CMS rule for 2014 attempts to address this problem by making patients who are in observation for more than 48 hours inpatients, as long as they meet inpatient criteria. Ross, who was part of a panel that met with CMS staff on behalf of the American College of Emergency Physicians (ACEP) during the open comment period for the proposed policy change, says one of CMS' goals with the new rule was to provide more clarification for providers.
"The rule is still a little bit confusing because it is not clear if it trumps or replaces InterQual criteria, or if the two [sets of criteria] work together," says Ross. "I think they are supposed to work together."
While CMS was not responsive to all the changes that Ross and the other ACEP representatives wanted to see in the new rule, he was pleased that the agency dropped wording to the effect that it would only pay for observation care beyond 48 hours if the care occurred in an inpatient bed. "That would have been disastrous for observation units," he says. "They took that wording out."
Ross suggests that the 48-hour rule would be unnecessary if all observation care was, in fact, provided in protocol-driven, dedicated observation units. "The length-of-stay for these units averages 15 hours," he says. "Less than 1% of patients are in observation units for more than 36 hours."
Ross notes that it is also quite clear that placing observation patients in beds on inpatient floors is far from ideal. "The average inpatient bed is made for a five-day patient. It does exceptionally well with five-day patients. But it has been shown time and again that these 6- to 24-hour patients fail miserably when you put them in an inpatient bed," he says.
Christopher Baugh, MD, MBA, medical director of the ED observation unit at Brigham and Women's Hospital in Boston, MA, favors the new CMS rule, but he anticipates that hospitalizations resulting from the new 48-hour rule will be among the most scrutinized by the Recovery Audit Contractors (RAC) when the new policy goes into effect. "What I am reading is that patients can't just be occupying a bed for two midnights to qualify as an inpatient. They will also have to have an inpatient order, and they will have to have some medical justification for being in the hospital," he says. "I presume the RAC is going to be looking very closely at folks with two midnight stays and then discharged home to make sure that the hospital wasn't dragging its feet on the discharges just so the visits would qualify as inpatient stays."
Extend emergency culture into observation units
Baugh has published extensively on observation care, including a recent editorial in the New England Journal of Medicine in which he urged CMS to go farther in reforming observation payment polices so that patients will be protected from excessive expenses and efficient care will be incentivized.3
Baugh echoes many of the same points as Ross, noting that all observation care is not of the same quality. For instance, he points to several reasons why a dedicated observation unit can deliver better care than the kind of observation that takes place on inpatient floors. "This is a dedicated space where all you do is observation, so it is not a hybrid area or an inpatient area where you are also doing observation," he says. "There are policies in place and leadership that is accountable for these policies. That is a really important part of being efficient and good at delivering observation care."
Further, in addition to having evidence-based protocols in place to direct the care that is delivered, Baugh notes that observation units work best when they are integrated with the ED. "That is where most of the patients who are going to be populating the observation unit are going to come from, so having that emergency medicine culture extend to the observation unit is very helpful," he says. "There are people who don't have emergency medicine training who can run an observation unit efficiently. It is not a requirement, but in my mind, it is very helpful to have [an emergency medicine] background, and that extends to the nurses as well as the physician leadership."
Baugh favors the concept of a closed unit, where the unit leaders are the people who control what happens in that unit and which patients are admitted to the unit. "Some places have what is called an open unit where primary care physicians in the community can directly put one of their patients into the unit and also run the plan of care according to whatever they see fit," he says. "What you end up with is a lot of variation because you have a lot of cooks in the kitchen, and a lot of providers who don't do this type of care very often. As a result, they are not as efficient as they could be."
Alternatively, when a smaller group of clinicians runs a tightly controlled unit, they become very good at providing observation care in a highly efficient manner, adds Baugh.
Make use of case management
Baugh notes that one concern that many observation unit directors have is that their definition of what works well in an observation unit patients who would benefit from protocol-driven care is not the same definition that CMS uses. "You have patients who the observation literature would say are not good candidates for an observation unit visit, but they are still being billed as observation," he explains.
Another concern is the ever-changing regulations governing how and when insurers will pay for observation services. "This does affect how we deliver observation care because no one wants to be delivering care that they are not going to be paid for," says Baugh. "It seems like every three or four years, CMS does significant change to their regulations."
For instance, prior to 2007, CMS only paid for observation stays pertaining to three different diagnoses: chest pain, asthma, or congestive heart failure (CHF). "If it was anything else, it didn't qualify for an observation payment," notes Baugh.
However, the regulations changed dramatically in 2007, broadening the scope of observation care so that it can apply to virtually any diagnosis as long as patients meet certain criteria. For example, patients must require further evaluation or testing to determine whether they need inpatient care, explains Baugh. "These are people who need further medical care, and there is real uncertainty and risk around what could be going on with them," he says. "Providers need that additional time to better decide if the patients need an inpatient stay or not."
Now, there is a new CMS rule capping the length of time that a patient can remain on observation at 48 hours, and it may take some time for providers to get used to the latest change, notes Baugh. "There is a lot of memory around how things were in the previous set of regulations, and it is just hard to adapt to the way you think about observation so frequently," he says. "I think there is initially going to be some confusion [about the new rule]."
However, Baugh also notes that the new rule offers an opportunity for case management to work with providers so that they get a better sense of whether patients should be on observation or be admitted as inpatients as early into the patient visit as possible. "This has potentially big ramifications on the setting of care for a patient because if providers decide early on that a patient should be admitted as an inpatient and should not be cared for in an observation unit, it takes that option off the table, and the sooner you know that the better," says Baugh. "Getting case management to be involved early on [in these decisions] seems to be a best practice, although it is very hard to implement because the ED is admitting patients 24 hours a day, and it is hard for case management to be involved at 3 o'clock in the morning in real time, but that is a logistical issue that is being worked on in a lot of places."
Consider advantages in efficiency, expertise
Matthew Lyon, MD, FACEP, director of the ED observation unit at Georgia Regents University Medical Center in Augusta, GA, doesn't think the new CMS rule will have much of an effect on the care that is provided in dedicated observation units because most of them have a 24-hour LOS limit. "If a patient reaches the 24-hour mark in our observation unit, we switch them over to inpatient status, providing the patient meets admission criteria," says Lyon. "Our average LOS is about 16 hours."
However, the rule change may well impact patients who are placed on observation in an inpatient bed. These would be patients who have diagnoses that do not fit with one of the protocols that is used to drive care in the dedicated unit, explains Lyons. "Our emergency observation unit runs by protocols, so a patient has to have a diagnosis of chest pain, sickle cell disease, asthma, or CHF [to be admitted to the observation unit], and the provider has to believe the patient will respond to treatment in less than 24 hours," he says.
Lyon acknowledges that the kind of observation care provided to patients when they are placed in inpatient beds is going to be different than the care provided on a dedicated unit. "That type of care is generally not protocol-driven, so it is not as efficient, and patients generally stay longer than 24 hours," he says.
There is no denying the financial burden that falls on patients with extended observation stays. However, proponents of observation care counter that the kind of protocol-driven care that is provided in observation units actually saves money. This is, in part, because the dedicated clinicians who work in these units become specialists in providing observation care. "You will set up guidelines that allow for more of a focus on LOS, so you aim for a rapid turnaround," says Lyon.
A patient can expect to stay 12 to 16 hours in an observation unit, whereas the same patient might spend twice as much time in the hospital if he or she is placed on observation in an inpatient bed, where the patient may only be evaluated once a day, says Lyon.
"Our patients are evaluated every one or two hours to see if they meet discharge criteria so they come into the observation unit at any time of the day and they leave at any time of the day," says Lyon. "We have about a third of our discharges between midnight and 7 a.m., so we have a lot of people being discharged in the middle of the night, which is not traditionally when patients are discharged from the hospital. However, because we are there and the patients are there, as long as they have transportation, those patients are ready to get out of the hospital as soon as they can."
In his research completed on the subject, Baugh suggests that the health care system could save more than $3 billion a year if all the hospitals that currently do not have dedicated observation units actually put them in place.3 "You can turn what would have been 30 hours on an inpatient floor into 15 hours on an observation unit," he explains. "If you aggregate those savings over many thousands of visits and most hospitals will have thousands of observation visits that means a real change in capacity in a hospital to be able to take care of patients."
Given the advantages that observation units offer, Baugh is surprised that more hospitals aren't quickly adopting them. "It is a lost opportunity to reduce a significant amount of inefficiency. And this isn't just a pure business-economics perspective. I think patients receive better care when they are on a protocol. You are less likely to miss something, and you are less likely to have unnecessary testing that poses a risk to the patient," he observes. "Being in the hospital is dangerous, particularly if you are there longer than you need to be, so if you can take a 30-hour hospital stay and turn that into 15 hours, that is 15 fewer hours of the patient being in the hospital where he or she could get the wrong medicine administered, have a fall, or pick up an infection."
Prepare for more observation care
Baugh views the new CMS rule as an incremental step in the right direction, but he would like to see higher payments for observation care that is carried out in a dedicated observation unit as opposed to on an inpatient floor. "That could play a role in incentivizing observation unit use, which to me is a best practice for managing these observation patients," he says. In a recent editorial in the New England Journal of Medicine, Baugh urged CMS to go farther in reforming observation payment polices so that patients will be protected from excessive expenses, and efficient care will be encouraged.4
Given that the typical size of an observation unit is 8 to 10 beds, Ross suggests that almost any hospital could establish such a unit. "You are not opening additional space, you are just partitioning," he says. "You are saying this area will be where the observation patients go with the understanding that [approximately] 20% of these patients will fail in this setting and have to be moved to an inpatient bed."
Whether hospitals establish dedicated units or not, most experts expect the use of observation to increase. Lyon agrees that this is the trend he is seeing. "EDs are used frequently for primary care, but we also have a much sicker population that needs more care than what can be delivered in a short period of time," he says. "So, observation or a short stay is a very good tool. And as our population ages, it is going to be used more."
1. Ross M, Aurora T, Graff L, et al. State of the art: Emergency department observation units. Critical Pathways in Cardiology 2012;11:128-138.
2. Sheehy A, Graf B, Gangireddy S, et al. Hospitalized but not admitted: Characteristics of patients with "observation status" at an academic medical center. JAMA Intern Med. 2013 [Epub ahead of print]
3. Baugh C, Venkatesh A, Hilton J, et al. Making greater use of dedicated hospital observation units for many short-stay patients could save $3.1 billion a year. Health Affairs 2012;31:2314-2323.
4. Baugh C, Schuur J. Observation care high-value or a cost-shifting loophole? N Engl J Med. 2013;369:302-305.