A Quality Improvement Initiative: Patient-Specific Factors Contributing to Extended PACU Length of Stay and Subsequent Cost

Value-based care focuses on improving quality and outcomes for patients while avoiding excessive cost. This is accomplished by changing the methods by which patients receive care and proactively combating problems rather than reactively responding to them.(1) However, minimizing cost is a small component of value-based care and alone can be counterproductive.(2)

Value is multidimensional and includes clinical outcomes, quality of life, complications, and emotional wellbeing.(3) Legislation such as the Medicare Access and Children’s Health Insurance Program (CHIP) Reauthorization Act of 2015 (MACRA), has created programs such as the Merit-based Incentive Payment System (MIPS) and the Value Modifier (VM) to promote quality value-based care.(4) In similar form, physician education programs, including the Choosing Wisely campaign, help physicians spearhead healthcare change, increasing quality and decreasing cost.(5)

The American Medical Association (AMA) assists physicians with programs like the STEPS Forward modules that offer practice improvement strategies and models for physicians to introduce a successful value-based care model.(6) The American Urology Association (AUA) has created an initiative that includes resources, such as the MIPS Toolkit, to break down the complex program components so physicians can more easily understand each component and how to improve their practice, quality reporting, patient measures, and quality scoring.(7)

Delayed patient care and unnecessary delays waste time and resources for both the patient and the hospital. The post-anesthesia care unit (PACU) has been studied to determine safe and efficient methods for improving patient care, decreasing financial waste, and increasing patient flow through the hospital. PACUs were developed with the simple objective of reducing postoperative patient morbidity and mortality. However, extended PACU length of stay (PLOS) can delay operating room (OR) availability, delay initiation of postoperative care pathways, increase frustration for patients and healthcare professionals, and increase cost.(8) Most studies do not focus on patient-specific factors for excessive PLOS.

Previous studies have predominately investigated postoperative factors contributing to PLOS.(9–11) However, no identified studies to date have investigated preoperative and intraoperative variables potentially contributing to PLOS and the associated costs. The primary objective of this study is to examine preoperative characteristics, intraoperative variables, and postoperative complications affecting PLOS and cost. We hypothesize that patients with longer PLOS will have more complications as well as increased encounter costs.

Methods

The University of Kentucky Institutional Review Board (IRB) approved this retrospective study, IRB protocol 42525. Local National Surgery Quality Improvement Program (NSQIP) data were used to identify cases by Current Procedure Technology (CPT) code including laparoscopic partial nephrectomy (50543), laparoscopic nephrectomy (50545,50546), open radical cystectomy with urinary diversion (51595), and robot-assisted laparoscopic radical prostatectomy (55866). Standard NSQIP variables were extracted. We consider these procedures major as they require longer operative times and hospital stays.

All included cases were performed at the University of Kentucky between Jan. 1, 2012, and Dec. 31, 2016. All cystectomies at the University of Kentucky were performed in an open manner and most patients received incontinent cutaneous diversion utilizing ileum. The DaVinci Surgical System was used for all radical prostatectomies.

Partial nephrectomies are performed based on provider and preference, including robotic, pure laparoscopic, hand-assisted laparoscopic, and open modalities. Radical nephrectomies are performed in a pure laparoscopic, hand-assisted laparoscopic, or open manner.

Postoperative clinical outcomes were analyzed and common complications were noted, including death within 30 days, superficial surgical site infection (SSI), organ/deep space SSI, wound disruption, Clostridium difficile infection, sepsis/septic shock, renal failure, renal insufficiency, transfusions, pneumonia, urinary tract infections (UTIs), unplanned intubations or mechanical ventilation requirements greater than 48 hours, cardiac events (arrest, myocardial infarction), cerebrovascular accidents, and deep vein thrombosis (DVT)/pulmonary embolism (PE).

The institutional accounting system (Allscripts EPSi, Chicago, IL) was used to obtain cost data. Costs acquired were specific for the hospital encounter when the index procedure took place. Total costs (TCs) encompassed the sum of all costs for the encounter. Direct costs (DCs) were defined as all costs tied to the patient, including OR time, anesthesia time, bed occupancy, supplies, and nursing time. Hospital reimbursement for the services rendered defined the net revenue (NR). Contribution margin (CM) was defined as NR less direct costs. Overnight patients were considered as such if their LOS extended past midnight in the PACU. Time in the PACU was obtained by examining the patient record.

Binary variables were analyzed using the Mann-Whitney U test. p-values were calculated at 3 levels: p<0.05, p<0.01, and p<0.001. Median and interquartile range (IQR) were reported for all cost and operation variables, and all cost values are reported in thousands, U.S. dollars. Statistical analysis was executed using SPSSTM version 22 (IBMTM Corp., Armonk, NY).

Results

We included 787 consecutive patients who underwent selected urologic procedures at our institution from 2012 to 2016 in the study. Seventy-two percent of the patients were male and 28 percent were female. MPLOS was five hours (IQR 3–8 hours).

Preoperative Characteristics

Preoperative factors that significantly affect PLOS included weight loss (defined as weight loss >10 percent in the previous six months), preoperative dyspnea, creatinine levels, blood urea nitrogen (BUN) levels, and bilirubin levels (see Table 1). Patients with weight loss greater than 10 percent had a PLOS of three hours (IQR 2–6 hours), p<0.05 (see Table 1). Patients with dyspnea and an elevated creatinine level had a PLOS of six hours (IQR 4–8 hours) and five hours (IQR 4–9 hours), p < 0.01. Patients with an elevated BUN or bilirubin had a PLOS of five hours (3–9 hours) and four hours (2–5.3 hours), p < 0.05.

Eighteen percent of patients stayed overnight in the PACU. Elevated BUN and the presence of a bleeding disorder arose as the only significant factors affecting OPLOS (see Table 1). Roughly 22 percent of patients with an elevated BUN stayed in the PACU overnight, p < 0.0286 (see Table 1).

Perioperative and Postoperative Variables

Perioperative variables significantly affecting PLOS included procedure and ASA class. ASA class I–II patients had a MPLOS of four hours (IQR 3–8 hours) compared to ASA class III patients who had a MPLOS of five hours (IQR 3–8 hours). ASA class IV–V had a MPLOS of 11 hours (IQR 6–26 hours). The PLOS between each of the groups was significant, p<0.01 (see Table 2).

OPLOS was also significantly different based on ASA class. Roughly 8 percent of ASA I–II had OPLOS compared to 9 percent of ASA III and 22 percent ASA IV–V patients, p<0.05. Of OPLOS patients, 37.1 percent were ASA I–II, while 59.1 percent were ASA III and 3.8 percent were ASA IV–V. The PLOS between each of the groups was significant, p<0.01 (see Table 2).

Operative times were categorized into <180 minutes, 181–240 minutes, 241–300 minutes, and more than 301 minutes. Duration in the OR had no significant effect on PLOS (see Table 2).

OPLOS did not influence postoperative complications (including DVT, PE, PNA, death within 30 days, cardiac arrest, infection, and UTI) compared to non-overnight PACU patients (see Table 3).

Overnight Vs Non-Overnight PLOS

Patients who experienced OPLOS had a median PLOS of 17 hours compared to non-overnight PLOS patients whose median PLOS was 3.5 hours, p<0.0001 (see Table 4). As illustrated in Figure 1, the median stay for OPLOS patients was roughly four times longer than non-overnight patients in every CPT category, excluding radical cystectomy. Factors associated with OPLOS are elevated BUN and ASA class, p<0.05 (see Tables 1 and 2).

Figure 1. Median PLOS vs OPLOS by procedure

Patients staying overnight in the PACU accumulated a median of $58,300 in hospital charges (IQR $42,500–$66,000) while non-overnight PACU patients required almost $7,000 less at $51,800 (IQR $40,000–$62,600), p<0.01 (see Table 5). Additionally, overnight PACU patients were associated with $11,200 (IQR $9,300–$12,700) in median direct costs and $5,400 (IQR $4,500–$6,200) in median indirect costs compared to the $10,500 (IQR $8,500–$12,400) in median direct costs and $5,100 (IQR $4,100–$6,000) in median indirect costs accumulated by non-overnight patients, p<0.05 (see Table 5). Intensive care unit (ICU) direct costs were $0, indicating these patients were not transferred to the ICU.

Discussion

This study aimed to examine preoperative characteristics, intraoperative variables, and postoperative complications affecting PLOS and resultant cost. Preoperative patients with dyspnea, creatinine levels above 1.3 mg/dL, BUN greater than 21 mg/dL, or high ASA class were associated with increased PLOS. Additionally, elevated ASA class and BUN were associated with increased OPLOS.

While we initially hypothesized OPLOS would be associated with increased complications, our data did not support this. OPLOS patients stayed roughly 14 hours longer than non-overnight PACU patients, a finding not entirely explained by patient-specific risk factors. This may suggest that non-patient factors, such as hospital capacity and bed management, are influencing OPLOS.

Few studies have analyzed preoperative patient characteristics that assist in predicting PLOS and OPLOS. A study by Gabriel, RA, et. al., at the University of California San Diego Healthcare System observed more than 4,000 patients, seeking to develop an equation to predict PLOS based on preoperative factors.(12) This study concluded that the most predictive factors for extended PLOS included case duration and the use of general anesthesia. Additionally, obstructive sleep apnea, hypertension, coronary artery disease, chronic kidney disease, and history of postoperative nausea/vomiting were associated with significantly increased PLOS.(12)

These data reinforce our findings that patients with renal insufficiency prior to surgery exhibit increased PLOS. However, this study does not focus on OPLOS, a critical component of our study, due to its significant effect on cost and patient value. Additionally, operative duration did not influence PLOS or OPLOS in our study.

Previous studies have associated preoperative pain as being predictive of increased PLOS.(13) Additionally, postoperative pain is associated with ASA class level, which was a significant predictor of PLOS in our study.(14) Future studies could investigate how postoperative pain and pain management techniques influence PLOS.

Hospital cost was also found to be significant for patients with an increased ASA class and OPLOS. OPLOS costs considerably more — a median cost of roughly $1,000 more per patient (see Table 5). At our institution, approximately 1,600 procedures are performed in the main OR per month. Extrapolating the 18 percent OPLOS to this population yields $3.5 million in extra cost per year attributed to extended PLOS.

This study emphasizes the importance of determining factors prolonging PLOS to improve the value of care delivered to the patient. PLOS and OPLOS cause delays in OR preparation and occupancy times, leading to congestion throughout the OR. These delays can result in cancellation of scheduled cases, diversion of emergent cases to other facilities, and potentially lost revenues for the hospital.(15)

This study has specifically demonstrated preoperative and intraoperative factors have little predictive outcome on PLOS in our institution. However, nonclinical components of the system are influencing the delay, such as bed management. Thus, further research should focus on postoperative factors to improve quality and value of care.

Several studies have examined factors that improve the efficiency of the PACU from a postoperative systems perspective, such as bed management and hospital efficiency.(9–11,13,14,16,17) One study specifically analyzed delays in discharge from the PACU and found 41 percent of delays resulted from understaffing of non-physician and non-nursing positions and 36 percent resulted from delays in assessment by an anesthesiologist.(9) In another study, delayed discharge from the PACU was related to lack of ward-specific beds in 52 percent of cases.(18)

Other research investigated factors including physician or nurse preoccupation, bed unavailability, anesthesia assessment, OR delays, and many others.(10,11,13,16,17) However, prior studies have identified that the use of nurse-specific discharge criteria for PACU patients can decrease PACU delay by 24 percent.(11) Adjustments in PACU staffing and scheduling based on PACU workload and flow can additionally decrease PACU delays and promote flow through postoperative care.(16) Decreased time and delay in the PACU could then, in theory, reduced cost and hospital LOS. Our study found few patient-specific reasons for excessive PLOS and thus, is in accordance with the literature cited above concerning non-patient-specific factors.

There are some specific limitations to our study. Analysis of patient outcomes included only one institution, therefore cost of PLOS and OPLOS will vary by institution. Additionally, multiple surgeons were included in the study and individual factors were not analyzed. Hospitals may have variable PACU capacity in proportion to their overall bed capacity and may cohort patients from surgical subspecialties differently, both of which can influence discharged from the PACU.

Conclusion

ASA class, elevated BUN and creatinine levels, and the presence of dyspnea are all preoperative factors associated with increased PLOS in patients undergoing major urologic surgery. OPLOS is associated with ASA class, BUN levels, and the presence of a bleeding disorder. However, these variables are unlikely causative for the excessive PLOS seen in those patients staying in the PACU past midnight. OPLOS is associated with significantly increased TC.

Additional studies into non-patient causes of prolonged PLOS, such as bed management and hospital efficiency, are needed.

References

1. Value-Based Care & Lower Costs. Cleveland Clinic. Accessed March 2, 2019. https://my.clevelandclinic.org/health/articles/15938-value-based-care

2. What Is Value in Health Care? | NEJM. Accessed March 2, 2019. www-nejm-org.ezproxy.uky.edu/doi/full/10.1056/NEJMp1011024

3. Porter ME, Teisberg EO. Redefining Health Care: Creating Value-Based Competition on Results. Harvard Business School Press; 2006.

4. Usa M. Value-Based Programs. Published June 25, 2018. Accessed July 7, 2018. www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/Value-Based-Programs/Value-Based-Programs.html

5. Levinson W, Kallewaard M, Bhatia RS, et al. “Choosing Wisely”: A Growing International Campaign. BMJ Qual Saf. 2015;24(2):167–74. doi:10.1136/bmjqs-2014-003821.

6. AMA Unveils Resources to Help Docs Navigate Value-Based Care Models | American Medical Association. Accessed July 7, 2018. www.ama-assn.org/content/ama-unveils-new-online-resources-help-physicians-navigate-value-based-care-models .

7. American Urological Association. 2018 Merit-based Incentive Payment System Toolkit. Accessed July 7, 2018. www.auanet.org/practice-resources/patient-safety-and-quality-of-care/2018-mips-toolkit.

8. Samad K, Khan M. Unplanned Prolonged Postanaesthesia Care Unit Length of Stay and Factors affecting it. J Pak Med Assoc. 2006;56(3):5.

9. Tessler M, Mitmaker L, Wahba R, Covert CR. Patient Flow in the Post Anesthesia Care Unit: an observational study. Canadian J of Anesth. 1999; 46(4):348-351.

10. Masilang C, Shioura M, Esperanzate N, Lobes I, Vu F. Cause and Effect of Patient’s Delayed Discharge from PACU In Relation to Customer Safety and Satisfaction. J Perianesth Nurs. 2015;30(4):e31. doi:10.1016/j.jopan.2015.05.084.

11. Brown I, Jellish WS, Kleinman B, et al. Use of Postanesthesia Discharge Criteria to Reduce Discharge Delays for Inpatients in the Postanesthesia Care Unit. J Clin Anesth. 2008;20(3):175–9. doi:10.1016/j.jclinane.2007.09.014.

12. Gabriel RA, Waterman RS, Kim J, Ohno-Machado L. A Predictive Model for Extended Postanesthesia Care Unit Length of Stay in Outpatient Surgeries. Anesth Analg. 2017;124(5):1529. doi:10.1213/ANE.0000000000001827.

13. Junger A, Klasen J, Benson M, et al. Factors Determining Length of Stay of Surgical Day-case Patients. Eur J Anaesthesiol EJA. 2001;18(5):314.

14. Ganter MT, Blumenthal S, Dübendorfer S, et al. The Length of Stay in the Post-Anaesthesia Care Unit Correlates with Pain Intensity, Nausea and Vomiting on Arrival. Perioper Med. 2014;3. doi:10.1186/s13741-014-0010–8.

15. Solving the Negative Impact of Congestion in the Postanesthesia Care Unit: A Cost of Opportunity Analysis. Clinicalkey. Accessed October 14, 2018. https://www-clinicalkey-com.ezproxy.uky.edu/#!/content/playContent/1-s2.0-S0022480416304942?returnurl=https:%2F%2Flinkinghub.elsevier.com%2Fretrieve%2Fpii%2FS0022480416304942%3Fshowall%3Dtrue&amp;referrer=https:%2F%2Fwww.ncbi.nlm.nih.gov%2F .

16. Ehrenfeld JM, Dexter F, Rothman BS, et al. Lack of Utility of a Decision Support System to Mitigate Delays in Admission from the Operating Room to the Postanesthesia Care Unit: Anesth Analg. 2013;117(6):1444–52. doi:10.1213/ANE.0b013e3182a8b0bd.

17. Dembo J. Complications Occurring in the Postanesthesia Care Unit: A Survey. J Oral Maxillofac Surg. 1992;50(12):1348. doi:10.1016/0278–91(92)90256-Y.

18. Cowie B, Corcoran P. Postanesthesia Care Unit Discharge Delay for Nonclinical Reasons. J Perianesth Nurs. 2012;27(6):393–8. doi:10.1016/j.jopan.2012.05.013.