SHM Converge

Survey offers a snapshot of nationwide COVID-19 discharge practices



Discharge practices for COVID-19 patients vary widely at the nation’s academic medical centers, but there are some areas of strong concordance, especially related to procedures for isolation and mitigating transmission of COVID-19.

Dr. S. Ryan Greysen chief of hospital medicine at the University of Pennsylvania, Philadelphia

Dr. S. Ryan Greysen

In addition, most sites use some form of clinical criteria to determine discharge readiness, S. Ryan Greysen, MD, MHS, SFHM, said on May 5 at SHM Converge, the annual conference of the Society of Hospital Medicine.

Those rank among the key findings from of a survey of 22 academic medical centers conducted by the Hospital Medicine Re-engineering Network (HOMERuN), which was launched in 2011 as a way to advance hospital medicine through rigorous research to improve the care of hospitalized patients.

“When COVID came and changed all of our lives, HOMERuN was well positioned to examine the state of practices in member hospitals, and we set out some key principles,” Dr. Greysen said. “First, we wanted to respect the challenges and needs of sites during this extraordinary time. We wanted to support speed and flexibility from our study design to get results to the front lines as quickly as possible. Therefore, we used lightweight research methods such as cross-sectional surveys, periodic evaluations, and we use the data to support operational needs. We have developed linkages to more granular datasets such as electronic health records, but our focus to date has been mostly on the frontline experience of hospitalists and gathering consensus around clinical practice, especially in the early stages of the pandemic.”

In March and April of 2020, Dr. Greysen and colleagues collected and analyzed any discharge protocols, policies, or other documents from 22 academic medical centers. From this they created a follow-up survey containing 21 different domains that was administered to the same institutions in May and June of 2020. “It’s not meant to be a completely comprehensive list, but these 21 domains were the themes we saw coming out of these discharge practice documents,” explained Dr. Greysen, chief of hospital medicine at the University of Pennsylvania, Philadelphia, which is one of the participating sites.

Next, the researchers used a concordance table to help them keep track of which institution responded in which way for which domain, and they bundled the discharge criteria into five higher order domains: procedures for isolation and mitigating transmission; clinical criteria for discharge; nonclinical/nonisolation issues; discharge to settings other than home, and postdischarge instructions, monitoring, and follow-up.

In the procedures for isolation and mitigating transmission domain, Dr. Greysen reported that the use of isolation guidelines was the area of greatest consensus in the study, with 19 of 22 sites (86%) citing the Centers for Disease Control and Prevention and 7 (32%) also citing state department of health guidance. “Specifically, most sites included the ability to socially isolate at home (until no longer necessary per CDC guidance) as part of the criteria,” he said. Most sites (73%) required use of personal protective equipment (PPE) in transportation from the hospital and 73% gave masks and other PPE for use at home.

Dr. Maralyssa A. Bann hospitalist at the University of Washington/Harborview Medical Center.

Dr. Maralyssa A. Bann

Session copresenter Maralyssa A. Bann, MD, a hospitalist at the University of Washington/Harborview Medical Center, Seattle, another participating site, pointed out that the institutions surveyed look to the CDC as being “the single source of truth on discharge practices,” specifically material for health care workers related to discharging COVID-19 patients. “Notable specific recent updates include the recommendation that meeting criteria for discontinuation of Transmission-Based Precautions is not a prerequisite for discharge from a health care facility,” Dr. Bann said. “Also, as of August 2020, use of symptom-based strategy for discontinuation of isolation precautions instead of repeat testing is recommended for most patients. This is a rapidly evolving area.”

Practices in the clinical criteria for discharge domain varied by site. Slightly more than one-quarter of sites (27%) gave little or no guidance by using terms like “use clinical judgment,” while 14% gave very specific detailed algorithms. “Most sites fell in between and gave some parameters, usually along the lines of symptom improvement, temperature, and oxygen requirement, but the criteria were variable,” Dr. Greysen said. “For example, in terms of temperature, many sites said that patients should be afebrile for a specific length of time, 24-72 hours, while other sites simply said afebrile at discharge.” Meanwhile, the following criteria for discharge were addressed by relatively few sites: lab criteria (36%), age (36%), high-risk comorbidities (32%), or ID consultation (18%).

In the nonclinical/nonisolation domain, 73% of sites assessed for level of support available, though this was variably defined. Slightly more than half (55%) specifically assessed activities of daily living or the presence of a caregiver to assist, while 18% reported addressing durable medical equipment such as beds and toilets and access to food or medication supplies in ways that were specific for COVID-19 patients.

In the discharge to settings other than home domain, 77% of sites addressed discharge to skilled nursing facilities, inpatient rehabilitation, or long-term care, although specific requirements were often set by the accepting facilities. In addition, 65% of sites gave specific guidance for patients experiencing unstable housing/homelessness, usually recommending a respite facility or similar, and 59% addressed congregate/shared living spaces such as assisted living facilities. “Often the strictest criteria [two negative COVID tests] were applied to discharge to these types of settings,” he said.

In the postdischarge instructions, monitoring, and follow-up domain, 73% of sites reported providing home monitoring and/or virtual follow-up care. Programs ranged from daily texting via SMS or patient portals, RN phone calls, home pulse oximeters, and/or thermometers. In addition, 55% of sites had created COVID-specific brochures, discharge instructions, and other materials to standardize content such as use of PPE, travel restrictions, social distancing, signs and symptoms to watch out for, and what to do if worsening clinically.

Dr. Bann predicted future trends on the heels of the HOMERuN survey, including the development of more evidence and consensus related to discharge criteria. “Clarity is needed specifically around hypoxemia at rest/on ambulation, as well as more flexible criteria for oxygen supplementation,” she said. “We also think there will be a considerable amount of growth in posthospitalization monitoring and support, in particular home-based and virtual/remote monitoring.”

HOMERuN is supported by the Gordon and Betty Moore Foundation, the AAMC, the Patient-Centered Outcomes Research Institute, the Clinical Data Research Networks, the Patient-Powered Research Networks, and Agency for Healthcare Research and Quality. Dr. Greysen and Dr. Bann reported having no financial disclosures.

Recommended Reading

Hospital vs. outpatient management comparable for elderly syncope patients
The Hospitalist
Don’t discount discharge planning during pandemic
The Hospitalist
Inpatient opioid administration associated with postdischarge opioid use
The Hospitalist
AMA discharge linked to increased readmissions, discontinuity of care
The Hospitalist
Predicting patient risk of medication-related harm
The Hospitalist
‘Hospital at home’ increases COVID capacity in large study
The Hospitalist
Adverse events occur in LTC residents transitioning from hospital to nursing home
The Hospitalist
CMS launches hospital-at-home program to free up hospital capacity
The Hospitalist
Higher 10-day mortality of lower-acuity patients during times of increased ED crowding
The Hospitalist
In-hospital mobility impairment in older MI patients predicts postdischarge functional decline
The Hospitalist
   Comments ()