Taking Action from Estimated Date of Discharge Analytics
June 1, 2022 7 minutes read
With so much data available for analysis, it can be challenging to know exactly how to use each piece to truly maximize its value. One of these bits of data case managers can and should unlock and apply is estimated date of discharge (EDD).
The EDD is an estimate given as early as possible, ideally on the first or second calendar day of hospitalization for a patient. Even for a shorter stay, the date should be recorded at least 24 to 48 hours before discharge. Recording the time of day for discharge is not necessary, but when the clinical team is discussing the patient, they may reference morning or afternoon.
The date itself is not set by the case manager alone.
“It’s important to think about we, not I, as in ‘not the case manager alone,’” says Jeff Echternach, MBA, AS, NRP, DSC, technology officer for the Center for Case Management. “A lot of hospitals don’t pin down only one person or entity to set the date; rather, it’s the multidisciplinary care team made up of physicians, frontline nurses, nurse managers, physical therapists, case managers, anyone attending daily rounds. During the one- to two-minute patient update, they might provide a frame of reference by offering the EDD.”
The EDD can change during the hospitalization. Echternach adds that, during rounds, providers may say, “As long as we don’t see this or that, Friday still makes sense.”
“With that information, the clinical team can react and provide input,” he notes. “Whether a social worker might say that a benefit is not in place, or a pharmacist responds that a certain drug is not yet available, this conversation shapes the date and can inform a more accurate date.”
Determining the EDD
While an EDD is not required, it is incredibly helpful. Determining the EDD can happen in many ways, and the date is fairly objective.
“Clinicians can load or predict the first EDD by looking at the patient’s anticipated DRG,” Echternach explains. “They can use concurrent coding or other influences to predict what kind of diagnosis this patient will ultimately come up with, knowing that there is always going to be a margin of error. Patients are people, so they’re not produced on an assembly line. Regardless of that margin, for hospitals especially, inputting an EDD is so helpful toward predicting future capacity and census. It’s better to go in knowing the error rate or margin of error, and not assume that the EDD will 100% come true. Sometimes, the date will change.”
Significance of the EDD
The EDD, a small snapshot of information, attempts to provide a reference point for when the patient will go home — and, surprisingly, that minute detail can prove significant later.
As a frame of reference for rounds, the EDD can ground discussions for the clinical team and help them consciously think about the length of stay for that patient. The information also can be used to predict census and capacity, especially when a powerful data program provides a clear report of aggregate data.
“Once the date is in the computer, the hospital case managers can leverage the information as a census and throughput tool,” Echternach says. “This was the capacity center or transfer center who is looking at compiled data and can know whether they’re short 20 beds or in an ‘out of capacity’ crisis. The data can also be used as a predictor for future capacity planning.”
‘Three Moments in Time’
Most computer systems or electronic medical records (EMRs) can analyze different dates entered for a patient. Echternach recommends case managers and other members of the team consider “three moments in time” when it comes to recording an EDD:
- The first EDD, recorded within 24 hours (optimally) or 48 hours of admission;
- The second EDD, recorded two days from the first EDD;
- The last EDD, recorded at least 24 hours before discharge.
“For example, with a pneumonia patient, let’s say it’s Monday and you put in the first EDD today, the first day of admission,” Echternach explains. “The team predicts a Friday discharge, but throughout the week, more information is received about the patient. Perhaps they’re doing much better and will probably discharge on Thursday instead. The second EDD can be put in to reflect that. If the patient goes home that Thursday, then we want to be able to look back in time and in an aggregate manner at the first EDD prediction that clinicians put in and assess how accurate the first and second EDDs were in comparison to when the patient actually left.”
These three discrete moments offer significant insight into how decision-making and analysis is conducted in the hospital. Echternach suggests the key to analyzing these data is to look at the level of accuracy and ask good questions to assess what might be going on. Helpful questions to ask include:
- Were you accurate to the same day of discharge?
- If not, how many days off were you?
- Did it come in as an underestimate or overestimate?
- Does the team tend to underestimate at the beginning or end of the stay?
- Was it more accurate two days out from admission or two days out from discharge?
- In general, do you note a higher accuracy with specific diagnoses or with certain clinical teams?
“It can all be summed up as ‘Were we accurate, yes or no, how far off were we, and was it an over- or underestimate?’” Echternach notes.
Many other aspects can be considered, such as the floor of the hospital, the clinical team, the diagnosis, the type of patient, how the patient arrived, and even the kind of user that documented the EDD — a nurse, social worker, or others.
“There are so many variable slices that can help the hospital see patterns with clusters of accuracy or inaccuracy,” Echternach says. “The important thing is to ask how accuracy can be improved here, or what might be driving this grouping of high or low accuracy. Is it how the disease progresses? Is it the team? When you zoom out and then back in with these analyses, you may find awesome practices that this date alone could give you an early indicator of. If case managers and their teams do this right, it can be close to a diagnostic indicator of opportunities.”
Help for the Case Manager
Using EDD data can help improve processes, make accurate predictions, and give clinicians amazing insights into the overall picture of the hospital’s well-being. Of course, a system must be in place to enable these data to be captured. Ideally, the system should allow the user to differentiate between each entry so insight can be gleaned from the subtle differences from the first to last entry. It is equally important the team knows how to use the system.
“Make sure there is adequate teaching or structure for this information to be put into system,” Echternach says. “It’s not fair to put in an EDD 25 minutes before the patient exits the building. The latest input will always look perfect, but we won’t learn from that. There should be a statement in the system to exclude any estimates that were put in on the same calendar date that the patient actually leaves, and don’t include patients that discharged within one calendar day. It’s wise to consider that exclusion or a similar exclusion to get the most out of the data.”
Most EMRs do a decent job of supporting case managers and their needs, especially when it comes to visualizing and stratifying the data managed by the system.
“It all comes down to getting the data out of the EMR and into a dashboard that allows you to look at dimensions and to measure the three moments in time,” Echternach explains. “As long as the intervals of time make sense, you should have usable data.”
Taking Action
It is wise to act on pockets of high and low accuracy, Echternach says. Ways to make improvements based on the levels of accuracy include:
- Offering feedback to those groups and discussing what may be contributing to high or low accuracy;
- Where possible, replicating high accuracy traits and improving low traits.
Hospitals can capture many trends by studying the data and using them to make adjustments to practice. This can often lead to a shorter length of stay.
“EDD can also be one of the more effective advanced warning systems for capacity, such as capacity threat levels or management levels based on the quantity of estimated discharges,” Echternach says.
Other times, following EDD data can help hospitals improve their processes and address areas of weakness.
“One system with which I worked found that they were wildly inaccurate with sepsis,” Echternach recalls. “But the hospital acknowledged that sepsis care was an opportunity for improvement — not just for case management, but as a global hospital opportunity. As they were becoming more efficient, they put a lot of standards around how they care for sepsis patients. We noticed pockets where, in specific nursing units, looking month by month, it was indicating a really bad couple of months with inaccuracy. It was hard to explain that, but still it was statistically significant.”
Ultimately, the hospital found when staffing was low and case management had to cover nursing care in addition to their case management functions, there was a noticeable dip in the quality of EDD accuracy. They found a similar correlation in length of stay on those units. All this information together enabled the hospital to see the overall picture and address these challenges.
Trends like this can be uncovered in any hospital system, and with the guidance of the case management team, improvements can take place that result in excellent care for patients, encouraging metrics for the hospital, and a more positive experience for the case management team, who can better anticipate their workload and census.
With so much data available for analysis, it can be challenging to know exactly how to use each piece to truly maximize its value. One of these bits of data case managers can and should unlock and apply is estimated date of discharge.You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content