A Model of Creating Inpatient Hospital Surge Capacity through Early Discharge

William Dunne

EXECUTIVE SUMMARY

Disasters and catastrophic mass casualty incidents continue to occur in the United States and around the globe. These events challenge their communities to preserve life and often require responders to go to extraordinary measure and utilize austere conditions to fight injury, illness, and disease. These communities are often stressed to prioritize planning activities with limited funds to improve resilience. Therefore, a focus on efficiency and utility for day to day activities that support broader flexibility is key.

One such area of preparedness is in health care systems. Hospitals, specifically, are often operating at high capacity to ensure efficiency and effectiveness. When catastrophic events in the community create large influx of emergency patients or a disaster strikes and impacts the physical infrastructure including health systems, these organizations must have capacity to support “patient surge.” In events such as the terror attacks of September 11th, 2001, Hurricane Katrina, Super storm Sandy, and the mass shooting in Las Vegas in October of 2017, hospitals had to make rapid decisions on how to manage the large influx or movement of patients while providing high quality medical care for existing as well as these new patients.

The construct of “surge capacity,” strategies to manage these large influxes of patients, have been discussed for many years in emergency management and health care spheres. Experts have offered many opportunities to address the challenges of surge which include the development of cached space or alternate care sites, patient evacuation/transfer, modification of standards of care, and triage/reverse triage including early discharge. They also described the key contributors of success which are defined as “System, Stuff, Staffing, and Space”[1] However, few of these strategies have been developed in detail or studied with academic/scientific rigor.

This research project dives into the concept of reverse triage and early discharge/transfer as a strategy to create surge capacity. Building on the work of a multidisciplinary team at UCLA Health who developed a structure assessment tool to safety predict patients who could be discharged from their acute care hospital bed or transferred to a lower level of care, this study compared the clinical judgment and the utility of this structured assessment tool in the event of a catastrophic disaster. Clinicians (nurses) were provided a mock scenario and asked to use their clinical judgment in phase one and use the structured assessment tool in phase two. The charts of patients were then reviewed ninety-six hours (four days) after the assessments were completed to determine if they had needed any predefined critical interventions or if they had been discharged. This process leveraged work[2] at Johns Hopkins University by Kelen and his colleagues who created benchmarks of critical interventions that should be performed on patients in this acute care setting.

Eight-seven (87) clinicians were consented to participate in the study and forty-two (42) assessments were completed in each phase (control and test). The data was analyzed to assess the prediction, validation, assessment appropriateness utilizing a Chi-Square and Fischer’s Exact Test with the statistical significance assessed for a p-value of < 0.05.

In the limited number of assessments, there was only significantly statistical difference in outcome data between the clinicians’ clinical judgment and their utilization of a structured tool. This existed in the predictive ability difference in the accuracy of predicting safety of discharge (p<0.05). The control group (clinician judgment) performed nearly as well in all areas as the treatment group (structured assessment). Both groups erred on the side of patient safety and only a small percentage of patients would have been discharged inappropriately.

The tool proved to have stronger sensitivity but weaker specificity than the clinician judgment alone. The positive predictive values of the tool were also fairly strong: PV+ of 97.77% and PV- of 66.63%. This means that it accurately predicted those that were safe to discharge but it was less accurate in predicting those who were unsafe to discharge.

The additional key finding in the test group was that when the tool/prediction was inaccurate, the clinicians erred on the side of caution and recommended the patient remain admitted (over-triage) in 4 of 5 (80%) cases. This is compared to 4 of 6 (66.67%) cases in the control group. In summary, this pilot study showed that the assessment tool was a slightly better, but not statistically significant, at predicting patients who could be safely discharged (without the need for critical intervention) or those who needed to remain admitted (in need of additional care/critical interventions).

While this data is promising, the study had a number of limitations. The study was originally designed to capture a larger number and wider diversity of clinicians. The study was also limited with a relatively small number of assessments which constricts statistical analysis and broader generalizability of the results. The patient populations were less diverse than desired which also limits the analysis and strength of the study.

This project serves as a first foray into research into this area and will initiate broader discourse and additional studies. Opportunities include replicating the study with greater numbers and diversity of clinicians, assessments, patients, and perhaps sites. Another significant consideration is engaging the electronic health record (EHR) in real time to evaluate current patient status and interventions in place to allow for more rapid evaluation and objective decision-making.

Ultimately, the goal in this field is to provide clinicians with stronger guidance vetted in scientific evidence and supported in ethical, legal, and moral context to make difficult decisions in the face of catastrophic medical disaster situations. This will lead to the stronger ability of healthcare organizations to handle large “patient surges” related to these disasters. In turn, our communities will become more resilient to both man-made and natural catastrophic events.

 

 

 

 

[1] Amy Kaji, Kristi L. Koenig, and Tareg Bey, “Surge Capacity for Healthcare Systems: A Conceptual Framework,” Academic Emergency Medicine 13, no. 11 (November 2006): 1157–59, doi:10.1197/j.aem.2006.06.032.; Bruce M Altevogt, Institute of Medicine (U.S.), and Forum on Medical and Public Health Preparedness for Catastrophic Events, Medical Surge Capacity Workshop Summary (Washington, D.C.: National Academies Press, 2010), http://site.ebrary.com/id/10379896. ; Jamil D. Bayram et al., “Critical Resources for Hospital Surge Capacity: An Expert Consensus Panel,” PLoS Currents 5 (October 7, 2013), doi:10.1371/currents.dis.67c1afe8d78ac2ab0ea52319eb119688. ; Dan Hanfling, Institute of Medicine (U.S.), and Committee on Guidance for Establishing Standards of Care for Use in Disaster Situations, Crisis Standards of Care: A Systems Framework for Catastrophic Disaster Response (Washington, D.C.: The National Academies Press, 2012).

[2] Gabor D. Kelen et al., “Inpatient Disposition Classification for the Creation of Hospital Surge Capacity: A Multiphase Study,” The Lancet 368, no. 9551 (2006): 1984–90, http://www.sciencedirect.com/science/article/pii/S0140673606698085. ; Gabor D. Kelen et al., “Creation of Surge Capacity by Early Discharge of Hospitalized Patients at Low Risk for Untoward Events,” Disaster Medicine and Public Health Preparedness 3, no. S1 (2009): S10–S16, http://journals.cambridge.org/abstract_S1935789300001981. ; Kelen, Gabor D., Lauren Sauer, Eben Clattenburg, Mithya Lewis-Newby, and James Fackler. 2015. “Pediatric Disposition Classification (Reverse Triage) System to Create Surge Capacity.” Disaster Medicine and Public Health Preparedness, March, 1–8. https://doi.org/10.1017/dmp.2015.27

No Comments

Post a Comment