In many cases, ED boarding of patients, who are waiting for inpatient beds, causes ED overcrowding. Inpatient beds are often not assigned in advance (held) for ED patients, despite the relative predictability of the number of admissions that will come from the ED. Lack of inpatient bed availability is caused by several factors, such as
The (lack of) availability of inpatient beds is a predictable phenomenon that is rarely driven by pure clinical issues. Rather, it is driven more by preference, scheduling, and staffing practices.
The daily census highs and lows often result from volume variations in the elective OR schedule coupled with the variable nature of ED admissions. These surges in elective OR volume fill the inpatient units on and after the peak surgery days (generally Monday, Tuesday, and Wednesday) and leave those beds unavailable for urgent or emergent patients, who may require inpatient specialty beds. ED patients requiring admission must then be placed “off-service” or must board in the ED waiting for a bed. This consequence of longer LOS in the hospital is often seen with orthopedic patients and patients in the intensive care unit (such as elective heart patients and patients undergoing complex neurosurgical or vascular procedures).
Placing patients in “off-service” beds has several negative ramifications. Patients who are placed in inappropriate beds (medicine patients in surgery beds, surgery patients in medicine beds, non-ICU patients in ICU beds) often
- Stay longer in the hospital (longer LOS)
- Have a higher risk of complications and adverse events
- Are less satisfied (lower patient satisfaction scores)
Patients placed in off-service beds are frequently taken care of by nurses who may not be as comfortable caring for that type of patient—or as trained to do so. For example, orthopedic nurses have specific specialty training in orthopedics. These nurses are comfortable with and prefer caring for orthopedic patients. Placing a patient with a small bowel obstruction or acute renal failure on an orthopedic nursing unit does a disservice to both the patient and the nursing staff. When an “any bed available” approach is adopted (any open bed is acceptable when holding patients in the ED or PACU), physicians are forced to round in multiple areas of the hospital, lengthening their rounding period and delaying the physicians' availability to the OR, the clinic, their practices, their personal lives, and so on.
Therefore, a primary goal of patient flow improvement solutions is to ensure that patients are placed in the right bed, at the right time, with the right nursing staff. This approach not only improves flow, but also improves quality, safety, comfort, and experience for the patients and the providers.
Delayed discharges are another common problem plaguing the inpatient units of the hospital, resulting in the “daily dump” of patients late in the day. The delays may be related to rounding patterns of physicians, inadequate case management, poor discharge planning, or under-reporting of available beds. Occasionally the delay is a symptom of overworked nurses who would rather keep a patient ready for discharge than to receive a new admission (that may or may not be appropriate for the receiving unit).
Untimely Physician Rounding Patterns
There are 2 issues: one obvious and the other more nuanced. Physicians are not necessarily available to discharge patients, when the patients are clinically ready. They may be engaged in surgery, rounding in another hospital, seeing patients in their offices. The result is that patients may unnecessarily occupy an inpatient bed, while another patient is waiting in the ED, ICU, PACU, and the like for that bed.
A less obvious cause of bed unavailability relates to the order in which physicians round on and discharge their patients. Physicians usually begin their patient rounds seeing the sickest patients first. These patients are often in the ICU, step-down units, or units with a higher nurse-to-patient ratio. In other words, these patients are already receiving the highest level of care in the hospital, that is, nursing, monitoring, and so forth. While it may seem counterintuitive, when the hospital's utilization is very high, seeing the sickest patient first may actually delay appropriate disposition for that patient and other patients who are waiting to receive care.
This delay is the result of a cascade or “domino effect.” It is of little value to discharge a patient from a high-acuity bed to an unavailable low-acuity bed. Examples include a patient in a(n)
- ICU bed ready to go to an unavailable telemetry bed
- Telemetry bed ready to go to a med-surg bed
When a lower acuity bed is unavailable, attempting to move patients to a unit requiring fewer resources will not succeed until that lower acuity bed becomes available. Simple changes in rounding sequence that focuses first on patients ready for discharge makes those inpatient beds available to patients waiting in the ED or higher acuity (critical care) units. Then rounding on the ICU patients after the lower acuity patients are discharged or transferred allows a more streamlined flow of patients through the hospital. Because of the positive cascade effect, critical care beds are available more quickly. This process potentially reduces LOS and improves satisfaction while maintaining the quality of patient care.
Discharge Planning/Case Management
Another reason for late discharges is insufficient discharge planning and case management. Rarely is discharge a surprise to either the physician or the nurses caring for the patient. Ironically, the most common question asked by patients on hearing that they will be admitted to the hospital from the ED is “When can I go home?” Unfortunately, that may be the last time until the day of discharge that “going home” is discussed with the patient. This lack of planning often increases the patient LOS and the number of late discharges.
Discharge planning for patients undergoing elective admissions, that is, surgical, cardiac ablation, and so on, should begin in the physician's office before the patient enters the hospital and should be reinforced by every provider that the patient sees throughout the encounter—preadmission testing, AM admission areas, anesthesia interview, postoperative visit by nurses and physicians—and every postprocedure day's progress note should contain consideration of discharge needs. Patients and their families want to be informed about and part of the treatment plan. Contrary to popular belief, discussing discharge with patients on a daily basis does not make patients feel rushed out of the hospital. In fact, it makes them feel that they are a part of the treatment team with the ability to make decisions and understand what their needs will be when they get home (discharge medications, home health, physical therapy, etc).
“Bed hiding” is perceived to be a common issue underlying an inpatient unit's lack of available staffed beds. Because nurses often feel short-staffed and overworked, there may be a tendency to prioritize those patients ready for discharge last, thus preventing a new admission on that nurse's shift. Unfortunately, this and other practices that delay discharge result in poor patient satisfaction and comments such as, “At 9 am, the doctor told me I could go home and it was 4 pm before they let me leave.” It also results in a delay for inpatients boarding in the ED or PACU who are waiting for an inpatient bed to become available.
Admitting a patient to an inpatient bed is no small task. It requires a great deal of time and effort on the part of the nurse to complete the necessary admission forms, educate the patient regarding the unit and the patient's admission needs, and answer questions, all while taking care of the other patients on that nurse's team. The tendency may be to wait until the next shift or “let the ED handle it” rather than discharge a patient and then receive another patient with all that the admission work entails. This problem can be averted with a(n)
- Dedicated admitting or discharge nurses
- Express admission unit
- Clarification that discharges are not the sole responsibility of the nursing staff
- Correct staffing of environmental services (EVS) staff
Automating the discharge process further expedites the discharge: Once the discharge order is received, the
- Discharge nurse provides instructions and prescriptions.
- Discharge transport is automatically notified.
- Environmental services personnel are automatically notified to clean the room.
- Environmental services (EVS) personnel notify the “bed board” that the room is available.
Many hospitals actually decrease the level of EVS staffing on the afternoon shift, precisely when the demands for EVS to address admissions and discharges (clean the beds and rooms) goes up, not down.
Staffing Level Determination
When considering the following day's staffed bed availability, most hospitals use the midnight census to determine staffing levels on the inpatient unit. With the trend toward more observation and short-stay hospitalizations, patients' hospital stays are shorter and an inpatient bed may actually turn over twice (or more) in a given 24-hour period. Therefore, the midnight census may give a false depiction of the productivity of the nursing staff and the complexity of patient care in today's inpatient units. Bed turns should be calculated using inpatient units in order to determine the real census for staffing and budgetary purposes. Staffing should then be based on the real patient census throughout the 24-hour period including the acuity of the patient population served.
A major component of variation in healthcare lies in the variation in demand by day of the week and the imbalance in the capacity to absorb this variation. The key forms of variation throughout the week are primarily related to unscheduled and scheduled admissions. A related component of this variation is that the weekend services in most hospitals are very different than those provided during weekdays, yet illness is a 24/7 problem. Pending patient procedures and treatments therefore build up over the weekend placing excessive demands on services near the beginning of the week. This “batching” of demand subsequently drives an imbalanced occupancy, which usually peaks Tuesday and Wednesday and does not decrease until the following weekend. This variation can be documented and the capacity can be matched to the demand as it occurs throughout the week, in essence forecasting the need for services.
This variation in demand coupled with decreased services on the weekend is further complicated by the system-induced variation of imbalanced OR scheduling. Some of the lack of OR capacity is due to working through the queue that has built up during the weekend. The remainder lies in the preference of surgeons to front-load their schedules early in the week. At first glance, it would seem this preference is primarily related to the surgeons' desire to have their weekend free; however, as noted below, this early week scheduling preference is actually grounded in some fundamental flow/capacity issues.
While the origins of the OR schedule imbalance are multifactorial, the impact on the hospital census is clear—the hospital census peaks midday on Tuesday, when it is frequently 10% to 15% higher than the weekend low. Hospitals that are not at peak bed capacity can flex staffing to accommodate this variation, but this practice is rare and impractical. A better approach, and the only approach for hospitals at capacity, is to
- “Smooth” the variation from all admission sources of admissions as much as possible.
- Look closely at the weekend demand, performing work the day that it is needed.
The operating room has a significant impact on the flow of patients throughout the hospital. The peaks and valleys typically seen in the elective surgery schedule drive the corresponding patterns in the inpatient census. During the peak days, usually early in the week, electively scheduled patients fill the inpatient units. As a result, when urgent or emergent patients come to the ED, these specialty beds may not be available.
Example: Smoothing OR Flow
The common practice of performing elective total joint replacements early in the week is illustrative. Thoughtful surgeons may wish to discharge the patient before the weekend when the ancillary services like physical therapy are usually less available. As a result, early in the week, the orthopedic inpatient unit is filled with patients who will require an average 3-day stay. When the hip-fracture patient presents to the ED on Tuesday evening requiring an orthopedic bed, there may not be an available bed. That patient must then be placed off-service or board in the ED waiting for an inpatient bed. In addition, the patient may be placed on the pending (add-on) surgical list without a scheduled procedure time. The patient may even be bumped day to day, increasing the risk of complications and unpaid avoidable days for the hospital. These fluctuations decrease the predictability of nurse and physician scheduling (Figures 40-1 and 40-2).
Elective admissions vs emergent daily admissions.
Variability in elective surgical volume vs urgent and emergent volume demonstrates the day-to-day variability in elective surgical volume vs the relatively stable urgent and emergent volume (less variability around the mean—random but more predictable).
Smoothing the flow of elective admissions and ensuring that adequate capacity is available for the demand (ie, beds for urgent and emergent patients) results in less variation, that is, smaller ranges between high and low volume. Smoothing admission flow also increases the capacity of both the OR and the inpatient units, a significant benefit to the ED.
The OR block schedule is typically based on the surgeons' utilization and preferences, without consideration of the flow of patients on the inpatient units. Smoothing the elective OR schedule incorporates the flow of inpatients, ensuring that they are predictably admitted and discharged to the appropriate inpatient units. This smoothing process makes staffing, rounding, and overall care more predictable, with shorter stays and improved satisfaction (Figures 40-3 and 40-4).
Inappropriate patient placement because of overcrowding. This day-to-day variability leads to inappropriate patient placement on the inpatient unit.
Reducing variability by smoothing (graphic representation of the reduction of variability after smoothing).
This type of smoothing process requires substantial collaboration between the hospital and the surgeons. Surgeons must be willing to change the days of the week or hours that they work. To facilitate this change, hospitals must provide the surgeons with relevant data related to patient placement, patient satisfaction, nursing overtime, and physician office issues.
For its part, the hospital must agree to provide the necessary ancillary support services on the weekend to ensure that the quality of care remains consistent throughout the patient's stay—weekends as well as weekdays. The hospital will have to adjust staff schedules, which may not require hiring more staff but rather smoothing the staff schedule. With services more consistently spread throughout the week, the hospital no longer has to hire additional physical therapists to handle the peak loads.
If the process is well designed and successful, surgeons will find that access to the hospital and the OR is improved and their quality of life is improved with these changes. The surgeons and hospital will want to trial these changes to ensure that the new process is an improvement. The collaboration required to accomplish this change serves as a foundation for improved physician hospital relationships along with improved efficiencies and quality.
Smoothing the elective schedule requires recent, relevant, reliable, and actionable data. The most frequent obstacle in this process is the review of the wrong data. For example, surgical scheduling and information systems are often based on the time from surgeon incision to dressing rather than the time from patient entry to departure. Wrong data results in an inaccurate representation of the true time that a patient occupies an operating room and an inaccurate calculation of average case duration. Inaccurate information leads to scheduling and block-allocation errors that further inhibit smoothing initiatives. When actual and scheduled case durations match, scheduling is more accurate and predictable, the flow of patients to their destination units is improved, and bed availability is enhanced (Figure 40-5).
Accuracy of actual vs scheduled case duration.
Results from smoothing the flow of elective admissions are compelling (Figures 40-6 and 40-7). Reducing fluctuation
- Opens more functional capacity in the OR and on inpatient units
- Places patients in the appropriate bed and unit, reducing LOS
- Improves patient and physician satisfaction
- Increases functional capacity by reducing the volume variations
- Improves patient flow through the ED
- Improves ED patient satisfaction
Improvement in patient satisfaction after smoothing volume.
Improved physician satisfaction after smoothing elective surgical volume.
In many hospitals, the ICU and telemetry units are throughput bottlenecks. Due to the typical surgical schedule (described earlier), most ICUs are near or at capacity at the beginning of the week. The ICU flow is also impeded by nonstandard admission and discharge criteria, leaving the patient LOS contingent on individual physician preferences and practice patterns. This nonstandard approach causes uncertain and often excessive LOS, limiting bed capacity. Bottlenecks in telemetry also impede the movement of patients out of the ICU and the ED, further limiting throughput in both units.
Closed ICUs managed by a single group of physicians are an effective strategy to increase efficiency, as this practice increases the likelihood of consistent admitting, management, and discharge practices. When closed units are impractical, utilization of standard protocols and patient care pathways can also achieve this consistency.
Lack of telemetry capacity is frequently the cause of inpatient bottlenecks (see section “Example: Determining the Underlying Constraint”). The root cause is usually inefficient processes that prolong the patient LOS, thereby consuming more telemetry resources than necessary. Typical causes of delays in telemetry availability include
- Physician rounding and discharge practice inconsistencies
- Insufficient availability of stress testing
- Excessive consulting requiring multiple sign-offs to discharge the patient
- Lack of standardized admission and discharge protocols
Frequently, hospitals also keep patients on telemetry units far longer than clinically indicated. Establishing clear criteria for instituting and discontinuing monitoring of otherwise stable patients should be part of every hospital's management scheme.
Unavailable Long-Term Care
A final, noteworthy cause of inpatient throughput bottlenecks involves “downstream” bottlenecks due to unavailability of long-term care and psychiatric care facilities outside of the hospital. Patients waiting for long-term care can utilize as many as 10% or more of the available inpatient beds. In such circumstances, the solution to hospital crowding may require focus outside of the walls of the hospital proper. In Canada, this downstream bottleneck is a major cause of hospital overcrowding, using as many as 30% of the inpatient hospital beds in certain cities.