1. McCann-Spry, Lisa
  2. Pelton, JoAnne
  3. Grandy, Glenda
  4. Newell, Dawn


Demand for hip and knee arthroplasty in the United States is rising rapidly. This is creating considerable strain on healthcare systems' institutional resources and finances. To reduce this strain, Spectrum Health in Grand Rapids, MI, developed a strategy to decrease length of stay for most primary hip and knee joint replacement patients. Four specific interventions were undertaken concurrently: (1) communication with providers, (2) modification of patient communications, (3) standardized risk assessment and prediction, and (4) physical therapy on POD (postoperative day) 0 (i.e., the day of surgery). Length of stay was reduced an average of 0.5 days per patient for primary hip and knee joint replacement surgeries, creating a positive financial outcome without negatively affecting quality and patient satisfaction. This demonstrated the ability of a large, high-volume joint replacement center to transform organizational culture and generate rapid, measureable change.


Article Content

Data on inpatient surgery from the United States' Centers for Disease Control and Prevention (CDC) indicate that 719,000 total knee arthroplasty (TKA) procedures and 332,000 total hip arthroplasty (THA) procedures were performed in 2010 (CDC, 2012a). This was at a cost of more than $11 billion in 1 year (CDC, 2012b). The diagnosis-related group code for hip and knee arthroplasty is the highest hospital inpatient short-stay cost for Medicare. By 2030, the demand for primary and revision hip replacements is projected to more than double, while the demand for primary and revision knee replacements is projected to increase by more than 600%. The total number of replacements that will be performed annually by 2030 is projected to be nearly 4.5 million (Kurtz, Ong, Lau, Mowat, & Halpern, 2007).


The demand for THA and TKA procedures is rising rapidly because of the aging of the population and the prevalence of arthritis. This creates considerable strain on healthcare systems from an institutional resource perspective, with a significant number of hospital beds and staff devoted to arthroplasty patient care. This demand also generates a large financial strain on individuals, payers, and the national economy. One strategy to reduce the overall cost and strain on the nation's healthcare system is to decrease hospital length of stay (LOS). When handled correctly, this strategy also can help healthcare organizations meet quality and patient satisfaction goals. Patients are expecting shorter lengths of stay and faster rehabilitation as outcomes from hip and knee replacement continue to improve.



Spectrum Health is a major, regional, not-for-profit healthcare system with headquarters in Grand Rapids, MI. It offers a full continuum of healthcare services through 12 hospitals, more than 181 service sites, skilled nursing care facilities, and a nationally recognized health plan. Spectrum Health currently has more than 23,000 employees, 1,300 physicians and advanced practice providers, and 2,300 active volunteers.


Within two of its Grand Rapids-based hospitals, Spectrum Health has five orthopaedic units with a total of 129 orthopaedic beds. Services are offered in general orthopaedics, joint replacement, spine, foot and ankle, hand, upper extremities, sports medicine, orthopaedic oncology, pediatric orthopaedics, and orthopaedic trauma care. Orthopedic Network News has listed the Center for Joint Replacement (CJR) at Spectrum Health in Grand Rapids as the nation's sixth largest provider of hip and knee replacements, with approximately 2,400 replacement surgeries being performed annually by 23 employed and independent joint replacement surgeons.


Spectrum Health's CJR is committed to continuously improving the quality of its services and generating excellent patient outcomes. In 2007, the CJR was one of the first 15 programs in the United States to earn disease-specific certification in hip and knee replacement through the Joint Commission; it has been continuously certified since then. The Joint Commission's disease-specific care certification is nationally recognized among joint replacement centers as a process for achieving excellence. Certification requires a facility to demonstrate compliance with consensus-based national standards, effective integration of evidence-based clinical practice guidelines to manage and optimize care, and an organized approach to performance measurement and performance improvement that includes data collection and analysis (Joint Commission, 2014).


Under Joint Commission requirements, a minimum of four performance improvement measures are selected by each organization and certification must be renewed every 2 years. For the 2013-2015 certification cycle, Spectrum Health's CJR chose the following four performance improvement measures: decreased LOS, fewer blood transfusions, fewer postoperative complications, and fewer postoperative readmissions. Selection of these measures was based on their influence on quality, outcomes, patient satisfaction, and financial impact on the organization. A collaborative multidisciplinary team that included quality improvement specialists, nurses, rehabilitation specialists, care managers, and physician leaders led efforts to monitor and improve quality outcomes in all of these measures.


LOS Reduction Initiative

In conjunction with acquiring and maintaining Joint Commission disease-specific certification for its hip and knee replacement programs, Spectrum Health established an institutional goal to lower the LOS for primary THA and TKA patients. Several studies have shown that multimodal approaches can be effective at reducing LOS (Ayalon et al., 2011; Jones et al., 2011; Kehlet, 2013; Khan, Ng, Gonzalez, Hale, & Turner-Stokes, 2008). Thus, a multimodal approach was chosen to implement best practices across multiple physician groups that would generate meaningful improvement in average LOS for primary joint replacement patients.


The CJR's performance improvement team implemented several strategies to decrease LOS. This team included the clinical nurse specialist for joint replacement, the nurse managers of the joint replacement units, two orthopaedic surgeon clinical advisors for joint replacement, a quality improvement specialist, the rehabilitation supervisor, the care management supervisor, staff nurses from the joint replacement units, and other intermittent participants as needed for consultation. With the Joint Commission's 2-year recertification period running from June 2013 to June 2015, the LOS project interventions began in August 2013; the last intervention was implemented in April 2014.


The Goal

Joint replacement LOS at the CJR had remained consistent at approximately 3.34 days for hip replacements and 3.94 days for knee replacements as of Q1 2013. The team realized that it would require an organizational change to influence this measure. The decision was made to use two-night hospital LOS as the initiative's goal. This decision was based on the Advisory Board's reporting tool, Crimson, which contains physician performance data; when compared with similar institutions, the LOS at Spectrum Health was longer than average. To accomplish this, four specific interventions were undertaken concurrently: (1) communication with providers, (2) modification of patient communications, (3) standardized risk assessment and prediction, and (4) physical therapy on POD (postoperative day) 0 (i.e., the day of surgery).


Provider Communication

In January 2014, a communication tool in the form of a letter (see Appendix A) was developed by the CJR's two orthopaedic surgeon clinical advisors and its nursing leadership. This letter explained the evidence supporting the LOS initiative and provided information about the implementation plan. It was sent to all of the CJR's participating joint replacement surgeons and was presented in person to their office staff members in March and April 2014. The letter was also presented to all mid-level providers, orthopaedic residents, internal medicine hospitalists, preoperative assessment clinic staff, preprocedure planning staff, preoperative surgical staff, post-acute care consultants, care managers, rehabilitation specialists, and inpatient nursing staff during this time. The CJR's orthopaedic surgeon clinical advisors followed up with providers who had questions or concerns and encouraged their participation in the initiative. In addition, there was individual follow-up by one of the orthopaedic surgeon clinical advisors with the few physicians who did not promptly follow the new standard.

Appendix A:. Provide... - Click to enlarge in new windowAppendix A:. Provider Letter

Patient Communication

In addition to communicating with providers, the existing preoperative joint replacement patient education class content was modified in December 2013 to reflect the new LOS recommendations. As each patient was admitted, nursing, rehabilitation, and care management staff utilized white boards in the patient rooms to visually communicate the discharge goal of POD 2. Also, scripting was developed to help guide discharge conversations each day of the hospital stay (see Appendix B). The overarching strategy was to ensure that anyone who spoke to the patient or family anytime during the surgical process would present a unified message and expectations for hospital LOS.

Appendix B:. Staff C... - Click to enlarge in new windowAppendix B:. Staff Conversation Script for LOS/Discharge

Risk Assessment and Prediction

Another strategy was the implementation of a standardized tool to help guide discussions around discharge planning before surgery. With permission from its original author, the Risk Assessment and Prediction Tool (RAPT) (see Appendix C) was given to all joint replacement patients attending a preoperative patient education class beginning in September 2013. The tool was designed to predict the likelihood that a patient would be able to return home after joint replacement surgery. It asked six questions, including the patient's age, gender, distance he or she was able to walk before surgery, use of a walking aid, use of community resources, and if he or she would have help at home after discharge. Each patient response was scored according to the following three categories:

Appendix C:. RAPT... - Click to enlarge in new windowAppendix C:. RAPT

* Scores of more than 9 = highest likelihood of discharge to home


* Scores ranging between 6 and 9 = likely to discharge to home


* Scores less than 6 = likely to need subacute rehabilitation after discharge


The RAPT was sent to patients to be completed at home before attending class, and it was scored as they arrived. Individual guided conversations were held to discuss patient scores during the discharge education portion of the class. In addition, patients with scores less than 6 were encouraged to discuss specific discharge planning needs and preferences after class to ensure that appropriate arrangements were made prior to surgery. A list of local subacute rehabilitation and home care agencies was provided to all patients, along with a frequently asked questions sheet to help patients explore options for facilities. Decisions made at the class were communicated to hospital care management staff for reference when the patient arrived at the inpatient unit postoperatively.


POD Zero Physical Therapy

The final and most influential intervention in decreasing LOS was the initiation of physical therapy on POD 0 for elective joint replacement patients. Rapid mobilization and patient education have been shown to decrease LOS (Ayalon et al., 2011; Jones et al., 2011; Kehlet, 2013; Khan et al., 2008; Tayrose et al., 2013). Raphael, Jaeger, and van Vlymen (2011) demonstrated that patients who began mobilization on the day of surgery reported significantly less pain and had a shorter LOS than those who began later.


Prior to the LOS reduction initiative, Spectrum Health had routinely staffed its Rehabilitation Department from 7 a.m. until 5 p.m. This schedule did not allow joint replacement patients to receive POD 0 evaluation and rehabilitation sessions because most joint replacement patients received spinal anesthesia. The effects of the spinal anesthesia typically did not resolve until late afternoon, after the rehabilitation staff had left for the day. In December 2013, the decision was made to pilot 12-hour shifts for the physical therapists working on the total joint orthopaedic units from Monday through Friday. Two therapists were staffed each day from 7:30 a.m. until 8:00 p.m. In addition, one rehabilitation technician was scheduled for a 12-hour shift each day, Monday through Friday. The goal was for every patient to receive at least one session of therapy on the day of surgery.


The pilot continued successfully for approximately 4 weeks. Following the 4-week pilot, the key stakeholders in the Rehabilitation Department discussed the outcomes and decided to continue with POD 0 evaluations. The pilot resulted in an increase in patient satisfaction, an increase in employee satisfaction, and a decrease in LOS. Therefore, POD 0 physical therapy was implemented as a permanent change in practice in early 2014.


Patient selection for POD 0 therapy was based on the time the patient arrived on the inpatient unit and the availability of the physical therapy team. The physical therapists attempted to evaluate every patient who arrived on the unit prior to 5 p.m. If the patient was able to complete an ankle pump on the nonoperative side and was able to feel pressure on the nonoperative leg and buttocks, physical therapy was initiated. If the patient was numb, the physical therapist made a subsequent attempt in the evening, as time allowed.


One barrier to accomplishing POD 0 rehabilitation soon became apparent: the use of femoral nerve blocks for pain control in knee replacement patients. These blocks often caused quadriceps muscle weakness, which resulted in the operative leg buckling with weight-bearing activities. This increased risk for patient falls and did not allow for ambulation until the block resolved, which could take up to 24 or more hours after surgery. In partnership with the Anesthesia Department, adductor canal blocks were trialed in January 2014. These blocks introduced the pain and numbing medication lower in the nerve, below the motor branch. This block proved to be successful, continuing to provide pain control without affecting muscle strength in the operative leg, and femoral nerve blocks were completely discontinued by June 2014.


To prevent pain from being a barrier to POD 0 ambulation, a three-phased multimodal pain order set (see Appendix D) was utilized to ensure that patients received adequate pain control throughout their hospital stay. Patients were categorized as opiate naive, opiate exposed, or opiate tolerant on the basis of their recent narcotic use. The pain medication dosing was adjusted accordingly. The plan included a combination of nonsteroidal anti-inflammatory medications, oral narcotics, short- and long-acting narcotics, intravenous narcotics, and other adjuvant medications. If patients had satisfactory pain control, they would participate in rehabilitation sessions.

Appendix D:. Pain Or... - Click to enlarge in new windowAppendix D:. Pain Order Set

Another barrier to POD 0 ambulation was indwelling urinary catheters. To address this, it was decided that catheters were to be removed on POD 0 as soon as the spinal anesthesia resolved. This encouraged patients to get out of bed to toilet, which increased their day of surgery activity. The nursing staff members were educated about this early catheter removal process in March 2014.



The first month in which a noticeable decrease in average LOS was observed was February 2014 for THA patients (see Figure 1); a decrease was noticed in January 2014 for TKA patients (see Figure 2). Both of these decreases occurred after the pilot implementation of POD 0 physical rehabilitation in December. Another downward shift in average LOS occurred in April 2014 for both THA and TKA patients after all providers and staff were fully educated about the new LOS expectations.

Figure 1 - Click to enlarge in new windowFigure 1. Average length of stay for total hip arthroplasty (THA) patients at Blodgett Hospital, Spectrum Health, from January 2013 to December 2014. LOS = length of stay.
Figure 2 - Click to enlarge in new windowFigure 2. Average length of stay for total knee arthroplasty (TKA) patients at Blodgett Hospital, Spectrum Health, from January 2013 to December 2014. LOS = length of stay.

Ten continuous months of LOS data below the previous 2013 average resulted in a process shift in average LOS for both THA and TKA patients (see Figures 1 and 2). Overall, data showed an average decrease in LOS for these patients of 0.5 days.


Complications and readmission rates were tracked as separate performance improvement measures. It was important to the initiative's team that these rates not be adversely affected by the decreasing LOS. Fortunately, there was no negative influence on these measures with the implementation of the interventions and decreasing LOS.


In relation to cost, the initiative team estimated that the average decrease of 0.5 days per patient generated a cost savings of approximately $400 per patient after all interventions were implemented. In 2014, the CJR performed 2,167 primary THA and TKA procedures at a lower LOS, which generated approximately $866,800 in savings.


In Q4 2014, 489 TKA and THA surgeries were completed at the Blodgett Hospital location of the CJR. Of those, 360 (73.6%) were evaluated for rehabilitation on POD 0 by the Rehabilitation Department. The patients who did not participate in POD 0 physical therapy did not meet the criteria for a rehabilitation session because they were not in their postoperative room by 5 p.m., were still numb at the time of reevaluation, or were not clinically stable to mobilize.


In a 1-month follow-up study conducted in June 2015, the team found that physical therapists were able to evaluate the patients an average of 2 hours sooner than before, and patients stated that they felt less pain on the subsequent day after receiving POD 0 therapy. Also, some patients were able to receive a second therapy session on the day of surgery. The CJR is currently evaluating 81% of the patients on POD 0, of which 38% are receiving twice-a-day sessions on POD 0.


Spectrum Health utilizes Hospital Consumer Assessment of Healthcare Providers and Systems surveys to measure patient satisfaction. The CJR has consistently exceeded expectations for the question, "would you recommend this hospital to your friends and family?" Decreasing the LOS did not result in a decrease in patients' likelihood to recommend the CJR.



A baseline LOS of greater than 3 days had been the standard for primary TKA and THA patients for at least a decade at Spectrum Health. All of the interventions in this initiative, in combination, were needed to successfully reduce the baseline LOS for those patients.


The Interventions

First, having the strong support of the orthopaedic surgeon clinical advisors was crucial to the success of this LOS initiative. Their ability to communicate the recommendations to their peers and hold them accountable, if necessary, was essential. In addition, supplying the other providers, nurses, care managers, and office staff with timely communication about the practice changes allowed them to also hold surgeons and other providers accountable to the new LOS goal.


Second, it was critical for the patients to receive consistent, caring communication about LOS expectations. From the surgeon's office, through the preoperative joint replacement class, and throughout the hospital stay, all providers communicated with the patients to ensure that they knew about the 2-day LOS expectation. Scripting allowed the nurses and other staff to consistently provide key points each day as the patients progressed. Use of a white board in each patient's room to display the expected discharge date created a visual cue for patients and their families to reference throughout the hospital stay.


Third, with more than 90% of primary THA and TKA patients attending preoperative patient education classes, the RAPT was valuable in helping identify early on those patients who might require a rehabilitation stay at discharge. This classification allowed patients and families to better plan for these stays before surgery so discharge was not delayed. In addition, the RAPT helped other patients recognize up front that they were very likely to be discharged directly to home after their 2-night hospital stay. This allowed them to plan ahead for assistance as they transitioned to home.


Finally, it was important to have surgeon, physician, nursing, and rehabilitation staff support for the POD 0 rehabilitation schedule changes because this was a significant change from previous practice. In addition, being able to identify barriers to success, such as the femoral nerve blocks, and having the support needed to modify pain control practices were critical to ensuring the continued progress of the LOS work. Because of the success of POD 0 rehabilitation, appropriate patients are now being seen twice on the day of surgery and have expressed increased satisfaction with their care.


Expansion of Initiative: Spine

On the basis of the apparent early success of the joint replacement LOS initiative-and while the full implementation of all of its interventions was still occurring-the decision was made in February 2014 to expand that work into the elective spine fusion population. This population also struggled with an LOS longer than national averages at Spectrum Health, and there was no consistency in LOS communications. Therefore, the potential benefits of addressing the same problem in this population were clear. A letter similar to the primary joint replacement provider letter was developed and distributed in March and April 2014 to all personnel involved. In addition, the preoperative spine class content was modified at that time to give a consistent LOS message to patients.


There was also discussion about the efficacy and feasibility of POD 0 rehabilitation for these patients. During this discussion, it became clear that the majority of the rehabilitation content for this population was education and that these patients tended to have decreased memory capabilities on the day of surgery due to general anesthesia. Therefore, the decision was made to have the nursing staff focus on POD 0 activity and ambulation and have the rehabilitation staff begin their patient training and education on the morning of POD 1. Education was given to all spine nursing staff on the expectations for activity on the day of and day after surgery for fusion patients.


As a result, average LOS for this patient population has also decreased by approximately one-half day per patient. Readmissions, complications, and patient satisfaction continue to be monitored for the spine patients. To date, those outcomes have not been negatively affected by the decreasing LOS. This experience with a different population-in this case, spine fusion patients-lends additional support to the potential for long-term success for reduced LOS for primary THA and TKA patients.



Decreasing the hospital LOS for primary THA and TKA patients has the potential to generate many benefits, including reduced costs, lower complication and readmission rates, greater employee efficiency, higher patient throughput, and greater patient satisfaction. However, this requires the cooperation of all providers involved and a willingness to change existing policies and approaches across hospital departments and external clinical practices. This LOS initiative has demonstrated that large, high-volume joint replacement centers can transform organizational culture and generate rapid, measureable change when all practitioners and administrators involved cooperate to implement a multimodal approach.



The authors thank the CJR's orthopaedic surgeon clinical advisors, Dr. Karl Roberts and Dr. Thomas Malvitz, for their tireless commitment to quality and their role in the success of the CJR's LOS work. They are role models and champions and have continually pushed innovations forward while reminding us that doing the right thing for the patient is our most important job.


The authors also thank the nursing and rehabilitation staff of the joint replacement units at Blodgett Hospital. Their positive attitude, willingness to embrace change to improve patient care, and dedication to patients are an inspiration every day.


Finally, the authors thank Beyond Words, Inc., for assistance in the editing and preparation of this manuscript. The authors maintained control over the direction and content of this article during its development. Although Beyond Words, Inc., supplied professional writing and editing services, this does not indicate its endorsement of, agreement with, or responsibility for the content of the article.




Ayalon O., Liu S., Flics S., Cahill J., Juliano K., Cornell C. N. (2011). A multimodal clinical pathway can reduce length of stay after total knee arthroplasty. Hospital for Special Surgery Journal, 7(1), 9-15. [Context Link]


Centers for Disease Control and Prevention. (2012a). CDC/NCHS National Hospital Discharge Survey 2010: Procedures by selected patient characteristics. Retrieved from[Context Link]


Centers for Disease Control and Prevention. (2012b). Cost of hospital discharges with common operating room procedures in nonfederal community hospitals by age and selected principal procedure: United States. Selected years 2000-2010. Retrieved from[Context Link]


Joint Commission. (2014). Standards for disease-specific care certification 2015. Oakbrook, IL: Author. [Context Link]


Jones S., Alnaib M., Kokkinakis M., Wilkinson M., St Clair Gibson A., Kader D. (2011). Pre-operative patient education reduces length of stay after knee joint arthroplasty. Annals of The Royal College of Surgeons of England, 93(1), 71-75. [Context Link]


Kehlet H. (2013). Fast-track hip and knee arthroplasty. Lancet, 381(9878), 1600-1602. [Context Link]


Khan F., Ng L., Gonzalez S., Hale T., Turner-Stokes L. (2008). Multidisciplinary rehabilitation programmes following joint replacement at the hip and knee in chronic arthropathy. Cochrane Database of Systematic Reviews, 2, CD004957. [Context Link]


Kurtz S., Ong K., Lau E., Mowat F., Halpern M. (2007). Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. Journal of Bone & Joint Surgery, American Volume, 89(4), 780-785. [Context Link]


Raphael M., Jaeger M., van Vlymen J. (2011). Easily adoptable total joint arthroplasty program allows discharge home in two days. Canadian Journal of Anaesthesia, 58, 902-910. [Context Link]


Tayrose G., Newman D., Slover J., Jaffe F., Hunter T., Bosco J. III (2013). Rapid mobilization decreases length-of-stay in joint replacement patients. Bulletin of the NYU Hospital for Joint Diseases, 71(3), 222-226. [Context Link]


For more than 135 additional continuing nursing education activities on orthopaedic nursing topics, go to