Length-of-Stay: An Inadequate Measure of Patient Flow
By: Matt Carroll
Numbers should be the backbone of every business decision, no matter the industry. Unfortunately, using the wrong number or too many numbers creates confusion and bewilderment.
Using only one number on which to base a decision can be misleading, particularly when this indicator is poorly selected or it is not used in conjunction with other key measures. On the other hand, using too many numbers can cloud decisions by creating a lack of focus, lack of timeliness, and, ultimately, a lack of confidence.
For example, if a team manager focuses only on a baseball player's batting average in his selection process, he will miss a lot of other valuable facets in the player's game. If he considered two baseball players with batting averages of .300 and .270, respectively, he may choose the player with the higher average. But we all know that there are other important statistics, including home runs, RBIs, stolen bases, and fielding percentages, that should guide his decision in selecting the best player for his team.
Then again, some information, such as home field batting average during the day, how the player fields balls hit to his left side, how he hits on grass fields, etc., is interesting, but not essential to the selection process. This is simply too much information.
Let's apply this same kind of thinking to the hospital setting, where managers are fixated on length-of-stay (LOS) as the only measure of patient flow. In fact, many hospital scorecards are inundated with LOS trends and ratios (e.g., LOS by DRG or LOS by physician, etc.). But this laser-like focus on LOS can actually harm the patient flow effort because it tends to shift attention to a smaller percentage of patients, instead of the large number of patients that represent the bulk of a hospital's daily activity.
Worse, LOS is a poor indicator of patient flow because it is an imprecise snapshot of conditions taken in the middle of the night. When it comes to patient flow, what's important is the time of discharge; in other words, hours count, not days. If a patient who could have been discharged at 11 a.m. actually left at 6 p.m. and the hospital was on ambulance diversion for three hours between 11 and 6, the LOS measure would not push this issue to the surface. But adding an indicator that tracked percentages of discharges by noon would.
From another perspective, patient flow can be hurt if the hospital staff is perpetually measured, rewarded, and criticized on LOS. Every hospital has a few outliers that dramatically impact LOS, and while these outliers might have economic consequences to the hospital and should certainly get attention, their impact pales in comparison to the total impact of lost patient market share. Outliers have great influence over the numbers: often their LOS is 5 to 10 times longer than the typical patient. But, they often represent less than 5% of the total census. In fact, losing just a few patients due to poor patient flow could easily negate a year of work on the outliers. As the CMO of a client hospital recently stated, "We are focused on the one barge stuck in the mud instead of trying to clear the main channel."
Making LOS a priority leads the care team staff to spend their mornings exhausting all options to minimize length-of-stay for the low number of "barge-stuck-in-the-mud" patients instead of spending their time on the mainstream or "easy" discharges. Thus the majority of patients wait while the minority of patients receives a disproportionate amount of time and attention. If the same care management team were to truly focus on patient flow, they would prioritize their activities to ensure that the non-outliers, the typical patients, are unencumbered through the discharge process. Unfortunately, NUMBERS don't guide this behavior.
When the primary indicator for driving patient flow is length-of-stay (LOS), a predictable pattern emerges: famine and feast. A dearth of available beds in the morning is followed by a mountain of beds late in the afternoon. Unfortunately, the demand for hospital beds peaks about four to six hours before they are available, and then remains constant into the early evening. When demand remains constant and supply is created later in the day, there will be long waits in the emergency room and PACU, physicians who can't find beds for their patients, and ambulance diversion. This results in inexplicable waiting, which leads to loss of revenue, loss of market share, damage to physicians relationships, loss of community support, and so on.
Imagine shifting patient flow to earlier in the day, without considering LOS. Now the care management focus moves from the difficult cases to the easy discharges early in the day. Patients leave at an even pace throughout the day, smoothing out the supply of beds and better meeting demand. There is no extra work added, only a change in the daily routine. Focus for the current day becomes the mainstream discharges in the morning, followed by the more difficult discharges later and then working on the anticipated discharges for the following day.
For example, if three beds can be freed up early in the morning by focusing on the mainstream, an additional 15 people can be moved through the Emergency Room, assuming one in five will be admitted. Small things like this can have a big impact on patient flow.
To implement this type of behavioral change, it is critical to develop multiple key indicators that track the performance of the "main channel" and keep the "barge" numbers at the end of scorecard and out of the spotlight.
When this happens, patients are seen in a timelier manner and receive the correct level of care. Physicians are able to place the right patient in the right bed and nurses have a happier patient who is stabilized more quickly. Administration is happy, because diversion is decreased or eliminated and patient satisfaction increases.
The unintended benefit of this type of behavioral change can be a decrease in length-of-stay without it being the sole indicator and focus of the care management staff. Working on patient flow CAN decrease length-of-stay, but focusing on length-of-stay CANNOT increase patient flow.
If you want to know more e-mail me at matt.carroll@usccg.com
