BACKGROUND/RATIONALE:
Older people with chronic illness are at greatest risk for severe COVID-19 outcomes. At the time of submission, reports of 23 of 81 (28%) SARS-CoV2 infected residents died in a 130-bed Washington state nursing home facility for an overall mortality of 18%. COVID symptoms have become increasingly diffuse in the past months including fever (37-98%), tachycardia (43%), tachypnea (17%), fatigue (70%) and dry cough (59%). These symptoms are biased based on the surveillance and disease detection approach and lesser symptoms mostly go untested. Yet, standard CLC infection control practices COVID-19 screening still centers on assessing daily temperatures. However, the utility of fever as an indicator has been debated for older adults. Small studies have reported that nursing home residents have a lower basal temperature than community dwelling patients and are less likely to mount a fever in response to infection. With infection control practices dependent on fever as a primary symptom, there is a critical need to understand the temperature response to COVID-19 specifically among CLC residents.

OBJECTIVE(S):
The long-term goal is to reduce transmission of COVID-19 in CLCs by increasing early detection. We will study the vital sign trend for CLC Veterans, individually, and as a ward, to determine whether it can effectively identify Veterans and their CLC units with and without COVID-19. Our central hypothesis is that CLC residents will often mount a sub-threshold vital sign response to COVID-19, thus delaying early infection control response and repeat testing. We hypothesize that sub-threshold and ward level metrics can be used to retest for COVID-19 and trigger infection control practices that may improve outcomes.

METHODS:
1. Characterize the association between abnormal vital signs in the weeks before and after testing COVID-19 positive criteria (as defined by RAPID) among Veterans who develop COVID. We hypothesize that CLC residents with COVID-19 will be more likely to have vital sign perturbations that do not meet the threshold for retesting. This analysis will directly feed into the operations protocol for retesting.

2.Characterize the trend in the average ward vital signs in the weeks prior to identification of a CLC resident diagnosed with COVID-19 without inclusion of this resident's data. We hypothesize that the average daily ward vital signs will be elevated in the week prior to the first COVID-19 diagnosis. This analysis will be incorporated into our vital sign monitoring system as an early warning system.

3.Describe the mortality, hospitalization, and ICU utilizations of asymptomatic CLC residents with COVID-19 positive test results. We hypothesize that asymptomatic COVID-19 positive Veterans will have worse outcomes relative to COVID-19 negative Veterans. This analysis will feed into the CLC testing and isolation strategies for outbreaks.

FINDINGS/RESULTS:
Our findings suggested that the CDC fever threshold for COVID-19 screening should be reconsidered. A single temperature screening is unlikely to accurately detect COVID-19 in nursing home residents. Results were presented to the CDC, who then updated temperature guidance for nursing homes less than one month later (Rudolph, et al., Journal of the American Medical Directors Association). Repeated temperature measurement with a patient-derived baseline could increase sensitivity for surveillance purposes when applied to a nursing home population. In addition, ongoing work will determine the vital sign trend for Veterans residing in community living centers (CLCs), individually, and as a ward, to determine whether it can effectively identify Veterans and their CLC units with and without COVID-19.

Key Study Findings:

1. Older CLC residents are less likely to have 'fever' as a symptom of COVID-19. 

2. The Centers for Disease Control changed the definition of fever for nursing home residents based on the findings of this study

3. In older nursing home residents, the residents baseline vital signs (temperature, pulse, and pulse oximetry) may be better at identifying deviations related to infection

IMPACT:
In response to the COVID-19 pandemic and VA CLC guidance, we have built a system to monitor vital signs in CLCs, identify trends, and present the data to front line CLC leaders to make decisions about infection control practices. After piloting in COVID-19 hotspots of New York and Florida in early March 2020, the system was deployed to all 134 CLCs on March 25, 2020. The deployment predated the first reported VA CLC COVID-19 case. In doing so, we have created a nationwide monitoring system for COVID-19 symptoms. Because all CLCs are linked by a common medical record, we can supplement the information with demographic, comorbidity, laboratory, and clinical data. As the developers of the system, the investigative team has developed a broad network of CLC professionals who are responding to the COVID-19 crisis. This network can be leveraged to learn about responses to the current pandemic and future infectious disease outbreaks.

The findings of this research provided the CDC with timely evidence to change the guidance surrounding fever in nursing home residents. At the initiation, the fever threshold for COVID-19 screening (38.0 deg C) was determined to be infrequently attained in CLC residents with COVID-19 positive tests. A single temperature screening is unlikely to accurately detect COVID-19 in nursing home residents. Results were presented to the CDC, who then updated temperature guidance for nursing homes less than one month later (Rudolph, et al., Journal of the American Medical Directors Association; McConeghy J of the American Geriatrics Society).

PUBLICATIONS:

Journal Articles

1. Lally MA, Tsoukas P, Halladay CW, O'Neill E, Gravenstein S, Rudolph JL. Metformin is Associated with Decreased 30-Day Mortality Among Nursing Home Residents Infected with SARS-CoV2. Journal of The American Medical Directors Association. 2021 Jan 1; 22(1):193-198. [view]
2. Rumery K, Seo A, Jiang L, Choudhary G, Shah NR, Rudolph JL, Wu WC, Erqou S. Outcomes of coronavirus disease-2019 among veterans with pre-existing diagnosis of heart failure. ESC heart failure. 2021 Jun 1; 8(3):2338-2344. [view]
3. Rudolph JL, Halladay CW, Barber M, McConeghy KW, Mor V, Nanda A, Gravenstein S. Temperature in Nursing Home Residents Systematically Tested for SARS-CoV-2. Journal of The American Medical Directors Association. 2020 Jul 1; 21(7):895-899.e1. [view]
4. McConeghy KW, White E, Panagiotou OA, Santostefano C, Halladay C, Feifer RA, Blackman C, Rudolph JL, Mor V, Gravenstein S. Temperature Screening for SARS-CoV-2 in Nursing Homes: Evidence from Two National Cohorts. Journal of the American Geriatrics Society. 2020 Dec 1; 68(12):2716-2720. [view]
5. Erqou S, Papaila A, Halladay C, Ge A, Liu MA, Jiang L, Lally M, Menon A, Shah NR, Miech E, Virani SS, Zullo AR, Shireman TI, Longenecker CT, Ross D, Sullivan JL, Wu WC, Rudolph JL. Variation in statin prescription among veterans with HIV and known atherosclerotic cardiovascular disease. American heart journal. 2022 Jul 1; 249:12-22. [view]