key: cord-1021679-cu5b9hsf
authors: Roberts, Pamela; Wertheimer, Jeffrey; Park, Eunice; Nuño, Miriam; Riggs, Richard
title: Identification of Functional Limitations and Discharge Destination in COVID-19 Patients
date: 2020-12-02
journal: Arch Phys Med Rehabil
DOI: 10.1016/j.apmr.2020.11.005
sha: 616118b4f1450af36bdd277934024e0ddfcbe5a9
doc_id: 1021679
cord_uid: cu5b9hsf

Objective The objectives of this study were to identify functional limitations in COVID-19 patients admitted to acute care hospitals, to evaluate functional limitations by demographic, medical and encounter characteristics, and to examine functional limitations in relation to discharge destination. Design and Setting This is a cross sectional, retrospective study of adult patients with COVID-19 who were discharged from two different types of hospitals (academic medical center and a community hospital) within one healthcare system between January 1 and April 30, 2020. Participants Patients were identified from the Cedars-Sinai COVID-19 data registry who had a new onset positive test for SARS-CoV-2. A total of 273 cases were identified, which included 230 cases that were discharged alive and 43 patients who expired and were excluded from the study sample. Interventions Not applicable Main Outcome Measures Functional limitations in COVID-19 patients in acute care hospitals and the predictors for discharge disposition. Results A total of 230 records were analyzed including demographic, encounter, medical, and functional variables. In a propensity-score matched cohort based on age and comorbidity, 88.2% had functional physical health deficits, 72.5% had functional mental health deficits, and 17.6% experienced sensory deficits. In the matched cohort, individuals discharged to an institution experienced greater physical (62.7% vs. 25.5%, p<.001) and mental health (49.0% vs. 23.5%, p=.006) deficits than patients discharged home. Marital status (OR 3.17, p=.011) and physical function deficits (OR 3.63, p=0.025) were associated with an increase odds ratio of discharge to an institution. Conclusions This research highlights that functional status is a strong predictor for discharge destination to an institution for COVID-19 patients. Patients who were older, in the acute care hospital longer, and with comorbidities were more likely to be discharged to an institution. Rehabilitation is a significant aspect of the healthcare system for these vulnerable patients. The challenges of adjusting the role of rehabilitation providers and systems during the pandemic needs further exploration. Moreover, additional research is needed to look more closely at the many facets and timing of functional status needs, to shed light in utilization of interdisciplinary rehabilitation services, and to guide providers and healthcare systems in facilitating optimal recovery and patient outcomes.

function deficits (OR 3.63, p=0.025) were associated with an increase odds ratio of discharge to 23 an institution. 24

Conclusions: This research highlights that functional status is a strong predictor for discharge 25 destination to an institution for COVID-19 patients. Patients who were older, in the acute care 26 hospital longer, and with comorbidities were more likely to be discharged to an institution. 27

Rehabilitation is a significant aspect of the healthcare system for these vulnerable patients. The 28 challenges of adjusting the role of rehabilitation providers and systems during the pandemic 29 needs further exploration. Moreover, additional research is needed to look more closely at the 30 many facets and timing of functional status needs, to shed light in utilization of interdisciplinary 31 rehabilitation services, and to guide providers and healthcare systems in facilitating optimal 32 recovery and patient outcomes. 33 J o u r n a l P r e -p r o o f Introduction 49

In a short span of time, a pandemic impacted the world as we know it. The medical 50 community characterized the face of the severe acute respiratory syndrome coronavirus (SARS-51

CoV-2) as ranging from asymptomatic and mild cases to severe symptoms resulting in high 52 morbidity and mortality 1 . Initial studies on COVID-19 described typical clinical manifestations 53 including fever, respiratory symptoms, diarrhea, myalgia, and fatigue 2-5 . In more severe cases, 54 thrombocytopenia, acute kidney injury, acute myocardial injury, liver damage, gastrointestinal 55 damage, and acute respiratory distress syndrome (ARDS) are often observed 6,7 . Further, as 56 individuals in our global population present with diverse symptoms secondary to COVID-19, 57 scientific endeavors have elucidated additional clinical manifestations of the illness, which 58 include central nervous system dysfunction 4,8,9 . Mao and colleagues 4 found that patients with 59 more severe cases had increased risk for neurologic manifestations; almost 37% (78 of the 214 60 patients) with COVID-19 presented with neurologic symptoms, including impaired 61 consciousness (14.8%), acute cerebrovascular diseases (5.7%), and skeletal muscle injury 62 (19.3%). In a scoping review of the literature, COVID-19 has manifested in symptoms such as 63 headache, dizziness, hypogeusia, anosmia, altered level of consciousness, acute cerebrovascular 64 events, seizures, and ataxia 9 . 65

Clinical observations and pathological studies have shown that COVID 19 can result in 66 significant dysfunction both acutely and subacutely 7 . Jianan 10 referenced that mild cases in an 67 acute setting were found to have sleep dysfunction (63.6%), poor exercise endurance (61.4%), 68 mild dyspnea (57.9%), anxiety (62.1%), fear (50.0%), and poor motivation (41.8%). Kiekens 69 and colleagues 11 summarize reports of post-ICU functional difficulties, including severe muscle 70 weakness and fatigue, joint stiffness, critical illness myopathy and neuropathy, dysphagia, 71 J o u r n a l P r e -p r o o f (neuro)psychological problems, and impaired functioning including gait and mobility, activities 72 of daily living, and work. Preliminary research has begun to conceptualize the long-term 73 impairment and dysfunctions resulting from the multi-faceted body structure damage and 74 deconditioning 10 . Jianan 10 noted that weakness, motor dysfunction with reduced mobility, and 75 co-morbidities exacerbated by COVID-19 may result in significant and chronic functional 76 deficits. 77

Ceravolo and colleagues 12 conducted a systematic review on rehabilitation needs due to 78 COVID-19 and found papers were based on previous literature and not on the current pandemic. 79

They concluded further updates are warranted to characterize the emerging disability in 80 COVID-19 survivors and the adverse effects of chronic disability in this population. 81

It is plausible that residual organ dysfunction (i.e., cardiopulmonary; neurologic 82 symptoms), debilitating weakness, motor dysfunction with reduced mobility, and co-morbidities 83 exacerbated by COVID-19 will result in significant functional deficits. With the experience of 84 improved and discharged patients, timely rehabilitation intervention may improve prognosis, The aims of this study were to identify functional limitations in COVID-19 patients  95   admitted to acute care hospitals, to evaluate functional limitations by demographic, medical and  96 encounter characteristics, and to examine functional limitations in relation to discharge 97 destination. Using the demographic, medical, encounter, and functional limitation variables, we 98 hypothesize that functional limitations will be significant in predicting discharge destination 99 using bivariate and multivariate predictive modeling. 100

Design and Cohort Identification 102 This is a cross sectional, retrospective study of adult patients with COVID-19 who were 103 discharged alive from two hospitals within one healthcare system. One hospital is an academic 104 medical center and the other hospital is a community hospital, both located in Los Angeles, 105

California. Participants included patients discharged between January 1, 2020-April 30, 2020. 106

Patients were identified from the data registry using the ICD-10 diagnosis and positive 107 COVID-19 lab test. A total of 273 cases were identified, which included 230 cases that were 108 discharged alive and 43 patients who expired and were excluded from the study sample. Cases 109 were included if they were age 18 or older and were confirmed as a new onset COVID-19 patient 110 from January 1, 2020 through April 30, 2020. Most of the patients were from the academic 111 medical center (85%) and were admitted from home through the emergency room (80.5%). The 112 other 19.5% were admitted from an assisted living or skilled nursing facility. Ethical approval 113 and oversight were granted by the Cedars-Sinai Institutional Review Board (IRB). 114

Data elements for the study were identified through the electronic health record (EHR), 115

Epic 15 , and were extracted using the COVID-19 Population Discovery (Pop Disco) application. included the presence or absence of rehabilitation therapy services. Therapy services were 130 defined by participation in physical therapy, occupational therapy or speech-language pathology. 131

Medical variables included comorbidities. Comorbidity burden was quantified using the 132 Elixhauser coding algorithms for Elixhauser index to aid in prediction 16 and number of 133 Elixhauser comorbidities (0, 1-2, 3, > 4). We selected the Elixhauser Comorbidity Index to 134 adjust for comorbidities as it has been shown to be superior to other metrics 17-20 . 135

To examine functional status, functional variables retrieved from the EHR were 136 categorized into Physical Health, Mental Health, and Sensory Function. Physical Health included 137 the presence or absence of self-care deficits, motor deficits, dysphagia/eating deficits, and 138 bladder management deficits. Mental Health included the presence or absence of cognitive 139 deficits, depression, anxiety, or psychosis. Sensory Function included the presence or absence of 140 J o u r n a l P r e -p r o o f sensation deficits, vision deficits, or hearing deficits. Whether a functional deficit was mentioned 141 more than once, on different days, and within different assessments determined the threshold for 142 whether a patient had a deficit in one of the domains and was included in the analyses. Hearing 143 was the only variable that was mentioned only one time and was therefore excluded. 144

To control for confounding disparities of age and comorbidities within the sample (Table  145 1), cohort matching was established to compare functional status between the discharge home 146 versus discharge institution groups. To focus on functional status for the COVID-19 population 147 and its impact on discharge destination, age and Elixhauser comorbidity index were used to 148 create a propensity score matched comparison from the complete study sample of 230 patients. 149

The matched cohort consisted of 102 patients, 51 in the discharge home group and 51 in the 150 discharge institution group. 151

Continuous variables were compared with a 2-sample t test or with a Wilcoxon rank sum 153 test for nonnormally distributed data. Categorical variables were compared with a Chi-square 154 test. Since the cohorts of individuals discharged to an institution and home differed at baseline 155 (Table 1) , we conducted a propensity score-matched analysis to compare the outcomes between 156 groups while accounting for imbalances in baseline risk. We developed a multivariable logistic 157 regression model to estimate the propensity score for discharge location as the dependent 158 variable. Age and comorbidity score were the independent variables in the model. Between-159 group imbalances were considered to be small if the absolute standardized difference for a given 160 covariate was less than 10%. Analyses were performed using SAS statistical software, version 

In analyzing the sample (n=230) by discharge destination of home versus discharge to an 166 institution, the age in years was 56.75 + 16.62 and 75.77 + 14.65, p<.001, respectively. As in the 167 overall sample, the majority in the discharge home versus institution were male (61.2%). For 168 ethnicity, non-Hispanic was greater in both the discharge home cohort (75.8%) and the discharge 169 institution cohort (84.6%). In the discharge home and institution cohorts, race also showed a 170 higher percentage of white, 70.3% and 61.5%, respectively; p=.003. More of the patients who 171

were discharged home were married or had a domestic partner (61.2%) as compared to those 172 discharged to an institution (40.0%), p=.003 (Table 1) . 173

Length of stay (LOS) for the patients who were discharged home was shorter, 7.66 + 5.87 175 days as compared to those who were discharged to an institution, 15.15 + 9.39 days, p<.0001. 176

For patients with an ICU stay, there was more variability in their discharge destination with 177 38.5% discharged to an institution and 23.0% home, p=.015 (Table 1) . Specifically, patients 178 who had an ICU stay and who were discharged to an institution spent 10.71 days on average in 179 the ICU as compared to those who were discharged home who spent 6.41 days on average in the 180 ICU. Ventilator use was higher in the discharge to institution cohort (53.8%, p<.0001). Overall, 181 rehabilitation therapy provision was 18.2% in the discharge to home cohort and 47.7% in the 182 discharge to institution cohort (Table 1) 

The higher Elixhauser Index was seen in the discharge to institution cohort 15.52 + 8.59 188 as compared to the discharge home cohort, 8.42 + 6.93; p<.0001. The percentage of Elixhauser 189 comorbidities was higher in the discharged to institution cohort, with 71% having > 4, when 190 compared to 27% of the discharged home cohort. Table 2 Mental health deficits yielded a significant difference between cohorts, with a higher percentage 208 of mental health deficits for those being discharged to an institution (49.0%) versus home 209 J o u r n a l P r e -p r o o f (23.5%), p=.006. Provision of rehabilitation therapy was generally not significant based on 210 discharge destination; however, those patients who received occupational therapy were more 211 likely to be discharged to an institution (27.5%) when compared to those discharged home 212 (7.8%), p=.009 (Table 4) . 213

The significant variables from the bivariate analysis were included in the logistic 214 regression model including physical health, mental health, marital status, and LOS. Marital status 215 of single/divorced/widowed was significant for discharge to an institution, OR=3.17, p=.011. 216

Presence of physical health deficits was also significantly associated with discharge destination 217 to an institution (OR=3.63, p=.025). Table 5 shows discharge disposition by type of therapy 218 service for both ICU and non-ICU patients. Patients who were in the ICU and who received 219 occupational or physical therapy were more likely to be discharged to an institution (80.0% and 220 59.2% for occupational therapy and physical therapy, respectively). confirmed COVID-19 who also have high comorbidity rates, it is prudent to integrate 299 rehabilitation therapies earlier in the hospitalization to address physical, mental, and sensory 300 deficits. The impacts of the pandemic will continue, and it is essential for the rehabilitation 301 community to optimize our ability to respond to the challenges and to determine the optimal 302 timing and dosage of rehabilitation services in our healthcare system. 

World Health Organization. Coronavirus disease 2019 (COVID-19) situation report-45

Clinical characteristics of coronavirus disease 2019 in China

Clinical features of patients infected with 312 2019 novel coronavirus in Wuhan

Neurologic Manifestations of Hospitalized 314

Patients with Coronavirus Disease 2019 in Wuhan, China. JAMA Neurol

Comorbidity Indices Versus Function as Potential Predictors of 30-Day 415

Readmission in Older Patients Following Postacute Rehabilitation

Comorbidities in Predicting Acute Care Readmissions in Medically Complex Patients

Effectiveness of early discharge planning in acutely ill or injured hospitalized 422 older adults: A systematic review and meta analysis

Effectiveness of nurse led early 424 discharge planning programmes for hospital inpatients with chronic disease or 425 rehabilitation needs: A systematic review and meta analysis

Rationale and principles of early 428 rehabilitation care after an acute injury or illness

Postacute Care Setting

A Systematic Review. Archives of Physical 431 Medicine and Rehabilitation

Functional 433 Impairment: An Unmeasured Marker of Medicare Costs for Postacute Care of Older 434

Rehabilitation and COVID-19: the Cochrane 437 Rehabilitation 2020 Rapid Living Systematic Review

Medicine and Rehabilitation and Pulmonary Rehabilitation for COVID-19