Objective
Many survivors of severe or critical COVID-19 have required rehabilitation during the pandemic. The primary objective was to compare characteristics and outcomes of survivors of severe or critical COVID-19 admitted to the inpatient rehabilitation facility during the first two waves of the pandemic. Our secondary objective was to identify the factors contributing to functional dependence on admission, discharge, and length of stay.
Design
This is a retrospective cohort study of 138 patients admitted to an inpatient rehabilitation facility in two waves after hospitalization for severe or critical COVID-19 illness between April 1, 2020, and May 3, 2021.
Results
Inpatient rehabilitation facility patients in wave 2 had significantly greater functional independence (GG scores) on admission (52; interquartile range, 44–58 vs 41; interquartile range, 28–52), lower incidence of dysphagia and anemia. The patients in both waves experienced similar functional improvement efficiencies with a median GG score change of 3.6 per day and similar discharge GG scores. Neurological sequela (odds ratio, 0.12; P < 0.001) and anemia (odds ratio, 1.35; P < 0.002) were identified as independent predictors of functional independence on admission.
Conclusions
Patients with functional deficits after COVID-19 should be considered for acute inpatient rehabilitation as both patient cohorts benefited from their inpatient rehabilitation facility stays with similar length of stays (11–12 days) and discharge to home rates (88%–90%).
Key Words: COVID-19, Rehabilitation, Functional Status, Neurological Manifestations
What Is Known
In March 2020, New York became the epicenter of the COVID-19 pandemic in the United states. Many patients throughout the pandemic have been discharged from acute medical surgical units to inpatient rehabilitation facilities (IRFs) after hospitalization for COVID-19. Research suggests that most COVID-19 IRF patients make progress comparable with other rehabilitation patients.
What Is New
Functional limitations after COVID-19 have not been re-evaluated in the subsequent waves of the pandemic. Patients in the second wave of admissions at this IRF showed less debility upon admission but still benefited from rehabilitation. Neurological complications and anemia were predictors of lower functional independence upon admission.
The COVID-19 pandemic, caused by severe acute respiratory syndrome coronavirus 2, is an ongoing outbreak of coronavirus disease 2019 (COVID-19) with multiple waves of infection.1 In March 2020, the first wave of COVID-19 began in New York State, which quickly became the epicenter of the COVID-19 pandemic in the United States.2 Nearly one million confirmed US cases of COVID-19 were reported by April 28, 2020, with more than 295,000 (approximately 30%) occurring in New York State alone.2 During the early stages of the outbreak, patients hospitalized with COVID-19 had poor prognosis, with two New York City hospitals reporting mortality rates of 43% among critically ill patients between March and April.3 Most early deaths in New York occurred in patients 50 yrs or older,3,4 most of whom had underlying health conditions including chronic cardiac and pulmonary diseases. In addition, many early, widely used COVID treatments and early ventilation strategies have been shown to have no efficacy in reducing disease duration or increasing survival.5–8
The United States experienced a second wave of COVID-19 cases beginning in June 2020 and a third wave beginning in September of the same year.9 While New York was largely spared from the second wave, cases in New York began to climb again in the fall of 2020. Despite an increase in the total number of cases in this new wave, the mortality rate was significantly reduced.10 According to the CDC, the death rate reached 800 deaths per day during the first COVID-19 wave in New York City, compared with a high of 98 deaths per day in the subsequent waves. Similar trends were observed in upstate New York and in other nearby states, including Massachusetts, Connecticut, and New Jersey.9 Other regions in the world have also reported reduced mortality during the second wave, including Spain11 and the United Kingdom.12 A number of factors may have contributed to this improvement, including evolving medical treatments, procedures, and vaccine administration.
The American Medical Rehabilitation Providers Association and the Ann Tumlinson Innovations Advisory reported that inpatient rehabilitation facilities (IRFs) have played vital roles during the COVID-19 pandemic in providing care to clinically complex patients.13 Patients admitted to IRFs in the United States are under the care of physicians and participate in intensive multidisciplinary therapy for a variety of conditions. During the pandemic, a New York State executive order prevented patients from returning to skilled nursing facilities while still testing positive for COVID-19, resulting in many patient discharges to IRFs. In the early wave of the pandemic, Richardson et al.4 reported 13.9% of discharged COVID-19 patients in the New York City area older than 65 yrs were sent to rehabilitation facilities or nursing homes. Debility requiring rehabilitation can result from COVID-19, with reports indicating weakness in both respiratory and limb muscles14 and difficulty walking independently after critical illness.14,15 Early studies have shown IRFs to be beneficial for COVID-19 patients, with rehabilitation progress comparable with patients of similar age and/or diagnosis.16–18 However, as treatments and medical practices surrounding COVID-19 evolve, so may rehabilitation needs.
In this retrospective study, we have compared the patient characteristics and outcomes of 138 COVID-19 patients admitted to an IRF after hospitalization for severe or critical COVID-19 during the first and second waves of the COVID-19 pandemic in New York State. In addition, we have identified the factors contributing to functional dependence on admission, discharge, and length of stay within the rehabilitation hospital.
METHODS
Participants
Between April 1, 2020, and May 3, 2021, 2343 patients were admitted to a single IRF and were screened as eligible (erehabdata.com). One hundred eighty-nine patients were identified as having COVID-19 using the International Classification of Diseases, Tenth Revision, code for COVID-19 (U07.1). Patients were included in the study if their rehabilitation admission followed an acute hospitalization with severe or critical COVID-19 as defined by the National Institutes of Health.19 These guidelines define severe illness as oxygen saturation (Spo2) less than 94%, a ratio of arterial partial pressure of oxygen to fraction of inspired oxygen (Pao2/Fio2) less than 300 mm Hg, respiratory frequency greater than 30 breaths per minute, or lung infiltrates greater than 50%. Critical illness is defined as respiratory failure, septic shock, and/or multiple organ dysfunction. One hundred thirty-eight patients met criteria and were included in the study. All patients were living home before their admission.
Ethical Considerations
This study was approved by St Peter’s Health Partner’s institutional review board (Albany, NY) in accordance with the Declaration of the World Medical Association and was found to meet criteria for an exempt study based on the federal regulations defined in 45 CFR 46.101(b). This decision included a waiver of informed consent/Health Insurance Portability and Accountability Act. This study conforms to all Strengthening the Reporting of Observational studies in Epidemiology guidelines. Required information is provided in the Supplementary Checklist (Supplemental Digital Content 1, http://links.lww.com/PHM/B729).
Demographic and Clinical Data
Patient demographics (age, sex, body mass index [BMI]), previous device use, previous living situation, comorbidities, pressure ulcers, and functional data were obtained from erehabdata.com, a database for inpatient rehabilitation outcomes. Comorbidities were identified using International Classification of Diseases, Tenth Revision, codes. Functional outcomes were derived from the GG Self-care and Mobility Activities items (Section GG0130 and GG0170 of the Centers for Medicare & Medicaid Services [CMS] issued IRF-PAI, Version 3.0) at admission and at discharge. The GG items (Section GG0130 and GG0170 of the CMS issued IRF-PAI, Version 3.0) are a standardized set of assessments used to evaluate and score functional abilities at admission and at discharge.20 The assessments are performed by registered nurses, occupational therapists, and physical therapists. Clinicians are trained in these assessments upon hire and are tested for competency every 2 yrs. Individual items are ranked on a scale of 1–6, with 6 representing complete functional independence for the activity. Self-care items include: eating, oral hygiene, toileting hygiene, showering/bathing, upper body dressing, lower body dressing, and putting on/taking off footwear. Mobility items include: rolling left to right, sit to lying, lying to sitting, sit to stand, chair/bed-to-chair transfers, toilet transfers, and car transfers. Walking items include: walking 10 feet, walking 50 feet with 2 turns, walking 150 feet, 1 step, 4 steps, 12 steps, and walking on uneven surfaces. The GG subscales on admission and discharge were calculated by erehabdata.com and are the sum of all individual item scores, excluding car transfers, 4 steps, 12 steps, and walking on uneven surfaces. Possible scores range from 18 to 108. The GG Self-care and Mobility scales show good internal consistency (Cronbach α ≥ 0.95) and substantial interrater reliability.21
The Brief Interview of Mental Status was used to assess cognition.22 This test is administered to patients within the first 3 days of IRF admission and assesses short-term memory and orientation. This assessment identifies patients with moderate or severe cognitive deficits and is included in Section C of the CMS issued IRF-PAI, Version 3.0.20 Total scores on the Brief Interview of Mental Status range from 0 to 15. According to test instructions, scores 13–15 indicate intact cognition, scores 8–12 indicate a moderate impairment, and scores 0–7 indicate a severe impairment. In cases where a patient is unable to complete the assessment, a score of 99 is assigned.
Information regarding the acute hospitalization for COVID-19 (hospitalization length, intensive care unit [ICU] vs ward, days after onset, COVID-19 medications administered, and need for invasive mechanical ventilation) was retrospectively obtained from the patient’s history and physical. Neurological complications included critical illness neuropathy and upper and lower limb paresis, encephalopathy, and delirium. Loss of taste or smell, dizziness, disorientation, amnesia, or other nonspecified signs or symptoms involving cognitive awareness were not included among the neurological complications.
Hematological data were obtained upon admission to the IRF and were collected retrospectively from the electronic medical record. Normoxia was defined as greater than or equal to 95% oxygen saturation. Elevated white blood cells were defined as greater than 10.8 × 109/l.23 Low red blood cells (RBCs) were defined as less than 4.5 × 106/l in men and less than 4.1 × 106/l in women.23 Low hemoglobin was defined as less than 14.0 g/dl in men and less than 12.0 g/dl in women.23 Low hematocrit was defined as less than 42% in men and less than 36% in women.23
The Functional Comorbidity Index described by Groll and To24 was used to identify comorbidities affecting functional outcomes. The comorbidities included in this index are arthritis, osteoporosis and/or fractures, degenerative disc diseases (including spinal stenosis and chronic back pain), asthma, chronic pulmonary diseases (eg, chronic obstructive pulmonary disease [COPD], emphysema), angina pectoris, past myocardial infarction, congestive heart failure or heart disease, neurological disease (eg, multiple sclerosis, neuropathy, Parkinson disease), cerebrovascular accident (including transient ischemic attack), peripheral vascular disease, diabetes mellitus type I or II, upper gastrointestinal disease (ulcer, hernia, reflux), obesity (BMI > 30), depression or other psychiatric diagnoses (eg, bipolar, schizophrenia), anxiety, visual impairments, and hearing impairments.
Statistical Analysis
Statistics were performed using SPSS (IBM) Statistical Software (V26). Continuous outcome variables were compared using Mann-Whitney U nonparametric test. Nonparametric statistics were chosen, because of the non-Gaussian distribution of the continuous variables upon visual inspection. Skewness and kurtosis Z scores were also calculated25 and found to exceed the recommended normality thresholds. Categorical variables were compared using Fisher exact tests. A Spearman correlation was used to determine variables associated with rehabilitation measures at admission and discharge. Variables were considered statistically significant if a P value was less than 0.01, to account for multiple comparisons. Missing data were excluded, and imputation was not used.
Using combined data across both waves, multivariable ordinal regression was used to determine variables independently associated with outcomes (GG admission scores, GG discharge scores, and length of stay). All variables demonstrating significant relationships with the outcome variables in the Spearman correlation were checked for multicollinearity, defined as a Spearman ρ greater than 0.8. All significant, unique variables were entered into the multivariable regression model, and nonsignificant variables eliminated individually using backward subtraction.
RESULTS
One hundred thirty-eight patients were admitted to the IRF between April 1, 2020, and May 3, 2021, after an acute hospitalization with severe or critical COVID-19 illness. Inpatient rehabilitation facility admissions occurred in two waves (black bars, Fig. 1) and coincided with New York statewide COVID-19 hospitalizations (gray line, Fig. 1). Wave 1 included 51 admissions to rehabilitation from April 8 to June 30, 2020, and wave 2 consisted of 87 admissions between August 2020 and May 2021. The demographics and comorbidities of the patients in this study are presented in Table 1. Compared with the first wave, the second wave patients were older (75 vs 64 yrs, P < 0.001). Chronic obstructive pulmonary disease incidence was significantly higher (Table 1; P = 0.006). While there was a tendency toward more preexisting conditions in wave 2, the overall Functional Comorbidity Index was not significantly different between waves (Table 1; P = 0.097).
FIGURE 1.
Patients hospitalized with COVID-19 at this inpatient rehabilitation facility and in New York State.
TABLE 1.
Comparison of patient demographics and comorbidities during wave 1 and wave 2
| Wave 1 | Wave 2 | P | |
|---|---|---|---|
| Demographics | |||
| No. patients | 51 | 87 | |
| Age, median (IQR) | 64 (58–76) | 75 (66–80) | <0.001a |
| BMI, median (IQR) | 26 (23–30) | 29 (25–35) | 0.019 |
| Sex (% females), n (%) | 22 (43.1%) | 42 (48.3%) | 0.599 |
| Rehabilitation impairment category | |||
| Stroke | 2 (4%) | 0 (0%) | 0.014 |
| Neurological | 5 (11.8%) | 1 (1.1%) | |
| Pulmonary | 1 (2%) | 6 (6.8%) | |
| Guillain-Barre | 1 (2%) | 0 (0%) | |
| Debility | 42 (82.4%) | 80 (92.0%) | |
| Comorbidities | |||
| Previous device use, n (%) | 7 (13.7%) | 20 (23%) | 0.177 |
| PVD/PAD, n (%) | 1 (2%) | 4 (4.6%) | 0.651 |
| COPD, n (%) | 2 (3.9%) | 19 (21.8%) | 0.006a |
| Chronic pulmonary disease (any), n (%) | 6 (11.8%) | 27 (31.0%) | 0.013 |
| Heart failure or disease, n (%) | 30 (39.2%) | 33 (37.9%) | 1.00 |
| Obesity (BMI > 30), n (%) | 13 (25.5%) | 39 (44.8%) | 0.029 |
| Upper GI disease, n (%) | 7 (13.7%) | 22 (25.3%) | 0.132 |
| Diabetes, n (%) | 27 (52.9%) | 43 (49.4%) | 0.727 |
| Depression/other mood disturbances, n (%) | 7 (13.7%) | 22 (25.3%) | 0.132 |
| Anxiety, n (%) | 16 (31.4%) | 32 (36.9%) | 0.581 |
| FCI, median (IQR) | 3 (2–3) | 3 (2–4) | 0.097 |
Median and IQRs of age and BMI are reported for patients admitted to the IRF during wave 1 and wave 2. The number and percent of females, patients requiring an assistive device (walker/wheelchair) before admission, and comorbidities during each wave are also included. Statistical significance was determined using the Mann-Whitney U test in the case of continuous variables, and Fisher exact test for categorical variables.
a P < 0.01.
FCI, Functional Comorbidity Index; GI, gastrointestinal; PAD, peripheral arterial disease; PVD, peripheral vascular disease.
In terms of their acute hospitalization, the first-wave patients were hospitalized longer (19 vs 14 days, P = 0.003; Table 2), were admitted to the ICU more frequently (82.4% vs 25.3%, P < 0.001), and were more frequently administered invasive mechanical ventilation (68.6% in wave 1 vs 17.2% in wave 2, P < 0.001; Table 2). Major pharmacologic treatments for COVID-19 during the first wave included hydroxychloroquine, which was administered to 80.4% of the wave 1 patients, and azithromycin (administered to 68.6% of the patients). The most prevalent COVID-19 treatments in the second wave were dexamethasone (75.7%) and remdesivir (75.9%; Table 2).
TABLE 2.
Acute hospitalization characteristics
| Wave 1 | Wave 2 | P | |
|---|---|---|---|
| Acute hospitalization | |||
| Length of stay, median (IQR) | 19 (11–33) | 14 (7–23) | 0.003a |
| ICU, n (%) | 42 (82.4%) | 22 (25.3%) | <0.001a |
| Ventilator, n (%) | 35 (68.6%) | 14 (17.2%) | <0.001a |
| COVID-19 treatmentsb | |||
| Hydroxychloroquine, n (%) | 41 (80.4%) | 0 (0%) | <0.001a |
| Azithromycin, n (%) | 35 (68.6%) | 6 (6.9%) | <0.001a |
| Dexamethasone, n (%) | 7 (13.7%) | 65 (75.7%) | <0.001a |
| Remdesivir, n (%) | 5 (9.8%) | 66 (75.9%) | <0.001a |
| Convalescent plasma, n (%) | 14 (27.5%) | 16 (18.4%) | 0.662 |
| Vaccinated (1 of 2 doses), n (%) | 0 (0) | 1 (1.1%) | 1.0 |
| Complications | |||
| DVT/PE, n (%) | 8 (15.7%) | 11 (12.6%) | 0.618 |
| Septic shock, n (%) | 18 (35.3%) | 13 (14.9%) | 0.010 |
| Multiorgan dysfunction, n (%) | 3 (5.9%) | 4 (4.6%) | 0.709 |
| Neurologic, n (%) | 13 (25.5%) | 9 (9.2%) | 0.014 |
Median and IQRs of the acute hospitalization lengths are reported for wave 1 and wave 2 patients. The number and percentage of patients requiring an ICU, ventilator, administered a COVID drug, or who experienced a complication related to their acute admission for COVID-19. Statistical significance was determined using the Mann-Whitney U test in the case of continuous variables, and Fisher exact test for categorical variables.
a P < 0.01.
b We were unable to obtain pharmacological treatment data for 6 patients (one wave 1 and five wave 2 patients).
DVT/PE, deep vein thrombosis/pulmonary embolism.
Patients admitted to the IRF had many functional needs during both waves of the pandemic. Dysphagia was present in 47.1% of wave 1 admissions as compared with 13.8% during wave 2 (P < 0.001; Table 3). A higher percentage of the patients required tube feeding in wave 1 (15.7% vs 2.3%, P = 0.005; Table 3). Activities of daily livings were also more impacted during the first wave, with GG scores significantly lower during the first wave of rehabilitation admissions (median GG score, 41; interquartile range [IQR], 28–52) as compared with the subsequent wave (median, 52; IQR, 44–58; Table 3). Most COVID patients (64.7%) admitted to the IRF were completely dependent for all walking items during the first wave, upon admission, as compared with complete dependence in walking for nearly 50% fewer (34.5%) of the second wave cohort (P < 0.001; Table 3).
TABLE 3.
Inpatient rehabilitation facility hospitalization characteristics
| Wave 1 | Wave 2 | P | |
|---|---|---|---|
| Admission to IRF | |||
| Functional needs | |||
| Dysphagia, n (%) | 24 (47.1%) | 12 (13.8%) | <0.001a |
| Swallowing regular food, n (%) | 28 (54.9%) | 75 (86.2%) | <0.001a |
| Swallowing modified food, n (%) | 16 (31.4%) | 11 (12.6%) | 0.013 |
| Tube/parenteral, n (%) | 8 (15.7%) | 2 (2.3%) | 0.005a |
| Pressure ulcers (all stages), n (%) | 19 (37.3%) | 19 (21.8%) | 0.075 |
| GG ADM, median (IQR) | 41 (28–52) | 52 (44–58) | <0.001a |
| Dependent for all self-care items, n (%) | 5 (9.8%) | 2 (2.3%) | 0.101 |
| Dependent for all mobility items, n (%) | 6 (11.8%) | 1 (1.1%) | 0.010 |
| Dependent for all walking items, n (%) | 33 (64.7%) | 30 (34.5%) | 0.001a |
| Discharge from IRF | |||
| Functional needs | |||
| GG DIS, median (IQR)b | 88 (74–96) | 94 (87–96) | 0.026 |
| Independent for all self-care items | 24 (51.1%) | 46 (60.5%) | 0.351 |
| Independent for all mobility items | 20 (42.6%) | 34 (44.7%) | 0.853 |
| Independent for all walking/stairs items | 9 (19.1%) | 14 (18.4%) | 0.999 |
| GG change per day, median (IQR) | 3.6 (2.5–4.8) | 3.6 (2.5–4.7) | 0.915 |
| LOS, median (IQR) | 12 (7–15) | 11 (7–14) | 0.347 |
| Discharge setting | |||
| Home, n (%) | 45 (90%) | 76 (88.4%) | 0.135 |
| Acute hospital, n (%) | 3 (6%) | 10 (11.6%) | |
| Subacute nursing facility, n (%) | 2 (4%) | 0 (0%) | |
| Died | 1 | 1 |
A comparison of the number and percentage of the patients with functional impairments upon admission and discharge from the IRF. A comparison of discharge settings is also shown. Statistical significance was determined using the Mann-Whitney U test in the case of continuous variables and Fisher exact test for categorical variables.
a P < 0.01.
b GG scores were not obtained for patients who died in the IRF or who were emergently discharged to a hospital.
DIS, discharge.
In terms of oxygen requirements, a significant number of the patients in both waves required tracheal or nasal cannula supplemental oxygen upon IRF admission. Interestingly, the patients were also more anemic during the first wave, with 90.2% of the patients with below normal hemoglobin levels as compared with 68.2% in the subsequent wave (P < 0.001; Supplementary Table 1, Supplemental Digital Content 2, http://links.lww.com/PHM/B730).
Despite wave 1 patients exhibiting greater functional dependence on admission, discharge GG scores were similar for both cohorts following rehabilitation. The median GG scores were slightly lower in the first wave (88 vs 94), but this did not reach the threshold for statistical significance (P = 0.026; Table 3). There was no difference in rehabilitation length of stay or GG efficiency between waves (Table 3). Despite improvement during the IRF stay, many patients still required assistance for various activities of daily living at discharge. Notably, only 18%–19% of COVID rehabilitation patients were completely independent for all GG walking and stairs items (Table 3). Most patients did not require oxygen supplementation at discharge (Supplementary Table 1, Supplemental Digital Content 2, http://links.lww.com/PHM/B730). The number of patients discharged on room air was slightly lower during wave 2 (77% vs 90.2%), but not significantly so, and may reflect the higher rates of COPD during this wave (Supplementary Table 1, Supplemental Digital Content 2, http://links.lww.com/PHM/B730). There was no difference in discharge destination during the two waves, with roughly 88%–90% of COVID patients discharged home. Two patients died while in rehabilitation, one occurring in each wave (Table 3).
Using combined data across both waves, variables associated with functional independence upon admission and discharge from the IRF were investigated in a series of zero-order Spearman correlations (Supplementary Table 2, Supplemental Digital Content 3, http://links.lww.com/PHM/B731). Although there was no relationship between age or BMI with GG scores on admission, the patients with a higher Functional Comorbidity Index had lower functional independence scores on admission (R = −0.256, P = 0.002; Supplementary Table 2, Supplemental Digital Content 3, http://links.lww.com/PHM/B731). Longer acute hospitalizations and ICU stays were associated with lower GG scores on admission as well (R = −0.277 and −0.341 respectively, P < 0.001 for both). Low hemoglobin (R = 0.354, P = 0.001 for hemoglobin), septic shock (R = −0.282, P < 0.001), and neurological sequela (R = −0.366, P < 0.001) were also negatively related to GG scores on admission (Supplementary Table 2, Supplemental Digital Content 3, http://links.lww.com/PHM/B731).
The variables most associated with GG scores on discharge were age (−0.280, P = 0.002) and GG admission scores (0.447, P < 0.001; Supplementary Table 2, Supplemental Digital Content 3, http://links.lww.com/PHM/B731). Only GG scores on admission (R = −.557, P < 0.001) and neurological sequela (R = 0.270, P = 0.001) were associated with length of stay (Supplementary Table 2, Supplemental Digital Content 3, http://links.lww.com/PHM/B731).
Multivariable ordinal regression was performed to determine the variables that independently predicted GG scores on admission. In a multivariable model, only neurological sequela (B = −2.24; 95% CI, −1.13 to −3.35; P < 0.001) and hemoglobin levels (B = 0.298; 95% CI, 0.11 to 0.48; P < 0.001) were independent predictors of functional independence on admission (Table 4). The median admission GG scores of the 22 patients with neurological complications was 38 (IQR, 25–49) compared with the patients without these complications (median admission GG, 59; IQR, 49–65).
TABLE 4.
Multivariable ordinal regression model
| B | Wald 95% CI | P | OR | |
|---|---|---|---|---|
| A. Independent variables predicting GG admission scores | ||||
| Neurological COVID-19 complications | −2.238 | −1.126 to −3.349 | <0.001a | 0.12 |
| Hemoglobin levels | 0.298 | 0.109 to 0.486 | 0.002a | 1.35 |
| B. Independent variables predicting GG discharge scores | ||||
| GG admission scores | 0.066 | 0.037 to 0.094 | <0.001a | 1.06 |
| C. Independent variables predicting length of IRF stay | ||||
| GG admission scores | −0.085 | −0.112 to −0.058 | <0.001a | 0.92 |
Variables independently associated with (A) GG admission scores (B) GG discharge scores, and (C) IRF length of stay. Only variables demonstrating significant relationships with the outcome variables are included.
a P < 0.01.
Multivariable models were also created to identify the strongest predictors of GG discharge scores and length of stay. In these models, only GG scores on admission significantly predicted these outcomes (Table 4).
DISCUSSION
In this study, we report the characteristics and functional needs of 138 COVID-19 patients requiring rehabilitation in New York during the first and second waves of the COVID-19 pandemic. Despite being older, the patients requiring rehabilitation in the second wave had shorter and less complicated acute hospitalizations. In addition, the patients in the second wave presented to the IRF with fewer functional limitations. Overall, the patients in both waves benefited from inpatient rehabilitation.
Rehabilitation needs of COVID-19 survivors are subject to change as both the severe acute respiratory syndrome coronavirus 2 virus and treatment strategies evolve. In this study, we have observed significant patient demographic differences between COVID-19 IRF admissions during the first few months of the pandemic as compared with later. Of interest, the median age of rehabilitation patients in the first wave was much younger—64 yrs compared with 75 yrs later in the pandemic. Other rehabilitation facilities have reported a similar age range (60–67 yrs)18,26–28 and prevalence of COPD (3.4%–21.4%) for patients during the first wave.26–28 At our facility, comorbidities, such as COPD and obesity, trended higher among survivors in the later cohort. These findings are interesting, given published findings that patients who are older have higher rates of mortality.29 It is likely that patients at higher risk of death did not survive the acute hospital course early in the pandemic, leading to fewer of these patients being transferred to the IRF. In subsequent waves, evolving therapeutic interventions became available, which may have allowed for more survivors. It is unlikely that patients were lost to subacute nursing facilities as an executive order (202.30) was put in place in New York State on May 10, 2020, preventing transfers of COVID-positive patients back to nursing homes.30
Length of acute hospitalization among the patients requiring rehabilitation was significantly longer during the first wave (19 vs 14 days). In addition, more patients were admitted to the ICU (82% vs 25%) and mechanical ventilation use was significantly higher (69% vs 17%). This was associated with more significant impairments during the first wave of IRF admissions, including dysphagia, and self-care, mobility, and ambulation dependencies. In this study, 64.7% patients were unable to walk without assistance upon admission to the IRF. Extremely low levels of functional independence for transferring and walking have been observed for rehabilitation patients28 and ICU survivors14 during the first wave. This is consistent with the finding that patients with critical illness and acute lung injury have long recovery times.31 While other rehabilitation facilities have reported similar percentages of ICU patients (50%–61%) during wave 1,18,26 to our knowledge, there are no studies documenting the rehabilitation needs of ICU and non-ICU survivors in subsequent waves. The COVID-19 patients in the second wave may have benefited from evolving pharmacologic treatment strategies, including the discontinuation of hydroxychloroquine and azithromycin in favor of remdesivir and dexamethasone, as currently recommended by the National Institutes of Health.32 As COVID-19 survivors later in the pandemic spend less time in acute care, their admission GG scores showed less debility on admission to IRF. In the current study, there were no patients in the IRF who had been fully vaccinated. However, it will be interesting to see how the COVID-19 vaccine impacts rehabilitation needs in the future.
Oropharyngeal and laryngeal dysfunction has been observed in COVID-19 survivors and is associated with prolonged intubation and ICU stays.33,34 In our study, 47.1% of the wave 1 patients had dysphagia. Other IRFs have reported rates of dysphagia as high as 85% among patients after severe and critical COVID-19.28 It is possible that the higher rates of ventilation may have contributed to the greater rates of dysphagia observed during the first wave.
Anemia has also been reported at high levels in patients with critical illnesses.35,36 Interestingly, hemolytic anemia has been documented in a small number of COVID-19 patients.37 Because of the retrospective nature of this study, we are unable to determine the cause of anemia in this population.
In this study, we found neurological complications and anemia most strongly predicted GG scores on admission across both waves. Similarly, Claflin et al.38 (2021) found that patients experiencing severe neurological complications during their acute hospitalization with COVID-19 experienced greater functional decline and greater frequency of dysphagia. In both this study, and study by Claflin et al.,39 only severe neurological complications were included. Other predictors of functional independence after COVID-19 that have been reported include age, respiratory failure, thromboembolic, and cardiac sequelae of the disease.39 Continuation of strategies aimed to minimize these complications should be considered.
During both the first and second waves, lower GG scores on admission were associated with lower GG scores at discharge and longer length of stays. Other factors including age, neurological complications, and a stay in the ICU were not found to be independent predictors of these outcomes. Both wave 1 and wave 2 cohorts experienced similar functional improvement efficiencies with a median GG score change of approximately 3.6 per day. As a comparison, the unweighted national average GG efficiency for rehabilitation patients with debility in 2019–2021 was 2.5–2.8 according to reports on erehabdata.com. Other reports have indicated similar rates of progress for COVID-19 rehabilitation patients compared with patients recovering from pneumonia or other illnesses.16,18 Both patient cohorts benefited from their IRF stays with similar rehabilitation length of stays (11–12 days) and discharge to home rates (88%–90%) for early versus late pandemic admissions. Other rehabilitation hospitals in the United States have reported average hospitalization lengths of 16–17 days for patients with COVID-19 during wave 1,17,28 suggesting that rehabilitation needs after severe or critical COVID-19 are variable. In the United States, length of stay in inpatient rehabilitation is largely determined by a patient’s insurer. Lower levels of functional independence and tier comorbidities, including COVID-19, contribute to higher case mix groups for Medicare patients.40 While length of stay is often longer for patients with a more severe case mix, physician/team recommendations as well as postdischarge care availability also guide length of stay.
Currently, there are limited data on the rehabilitation needs after wave 1. Most studies have examined rehabilitation needs during the first few months of the pandemic.16–18,26–28 Additional research at other facilities will be important in determining rehabilitation needs as both the virus and treatment strategies evolve.
Both groups saw significant improvement in patient’s respiratory status as the percentage of patients on room air at discharge increased in both groups. These findings continue to suggest the IRF as an appropriate setting that efficiently improved patients’ respiratory and functional status allowing most patients to transition home after critical illness.
LIMITATIONS
This study represents the findings of a single rehabilitation facility. There are likely some regional factors, including the surge of COVID-19 patients in New York City during the early phases of the pandemic, regional differences in treatment strategies, as well as New York State executive orders influencing how and where patients progressed from their acute hospitalizations. While New York fared better during the second and third waves of the pandemic, many areas of the United States experienced their peak daily mortality rate during this time. There continues to be a need for better treatment and continued education regarding vaccination.
In addition, we were not able to obtain the variants of SARS-CoV-2 that led to COVID-19 in these patients. It is possible that differences in the variants of SARS-CoV-2 strains may have contributed to the results. Because the retrospective nature of this study, misclassification bias may exist as well. The number of participants in our study is modest, and power of our analysis may allow for small differences to go undetected.
CONCLUSIONS
Throughout both the first and second waves of the COVID-19 pandemic, many survivors of severe and critical COVID-19 required rehabilitation. While patients in the early stage of the pandemic demonstrated higher levels of functional weaknesses, patients in later waves continue to participate and benefit from rehabilitation. Patients with functional deficits resulting from COVID-19 can require multiple therapy disciplines and should be considered for acute inpatient rehabilitation. Neurological complications and anemia were independently associated with low GG scores on admission to the IRF. Continued improvement of strategies to prevent and treat COVID-19 will likely impact rehabilitation needs in the future.
Footnotes
Sergo Gabunia and Erin Y. Harmon contributed equally to this manuscript.
The study was supported by James A. Eddy Research Institute, Sunnyview Rehabilitation Hospital, 1270 Belmont Ave, Schenectady, NY 12308.
Sergo Gabunia is in training.
Financial disclosure statements have been obtained, and no conflicts of interest have been reported by the authors or by any individuals in control of the content of this article
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.ajpmr.com).
Contributor Information
Sergo Gabunia, Email: dr.sergo@mail.com.
Erin Y. Harmon, Email: Erin.Harmon@sphp.com.
Matthew B. Sonagere, Email: Matthew.Sonagere@sphp.com.
Amy E. Teale, Email: Amy.Teale@sphp.com.
REFERENCES
- 1.Cacciapaglia G, Cot C, Sannino F: Multiwave pandemic dynamics explained: how to tame the next wave of infectious diseases. Sci Rep 2021;11:6638. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Gonzalez-Reiche AS Hernandez MM Sullivan MJ, et al. : Introductions and early spread of SARS-CoV-2 in the New York City area. Science 2020;369:297–301 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Chilimuri S Sun H Alemam A, et al. : Predictors of mortality in adults admitted with COVID-19: retrospective cohort study from New York City. West J Emerg Med 2020;21:779–84 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Richardson S Hirsch JS Narasimhan M, et al. : Presenting characteristics, comorbidities, and outcomes among 5700 patients hospitalized with COVID-19 in the New York City area. JAMA 2020;323:2052–9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Cummings MJ Baldwin MR Abrams D, et al. : Epidemiology, clinical course, and outcomes of critically ill adults with COVID-19 in New York City: a prospective cohort study. Lancet 2020;395:1763–70 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Oldenburg CE Pinsky BA Brogdon J, et al. : Effect of oral azithromycin vs placebo on COVID-19 symptoms in outpatients with SARS-CoV-2 infection: a randomized clinical trial. JAMA 2021;326:490–8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.WHO Solidarity Trial Consortium, Pan H Peto R, et al. : Repurposed antiviral drugs for Covid-19—Interim WHO Solidarity Trial results. N Engl J Med 2021;384:497–511 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Papoutsi E Giannakoulis VG Xourgia E, et al. : Effect of timing of intubation on clinical outcomes of critically ill patients with COVID-19: a systematic review and meta-analysis of non-randomized cohort studies. Crit Care 2021;25:121. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.CDC : COVID Data Tracker: trends in number of COVID-19 cases and deaths in the US reported to CDC, by state/territory. Available at: https://covid.cdc.gov/covid-data-tracker/#trends_dailycases. Accessed September 10, 2021
- 10.Hoogenboom WS Pham A Anand H, et al. : Clinical characteristics of the first and second COVID-19 waves in the Bronx, New York: a retrospective cohort study. Lancet Reg Health Am 2021;3:100041. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Rodriguez-Nunez N Gude F Lama A, et al. : Health indicators in hospitalized patients with SARS-CoV-2 pneumonia: a comparison between the first and second wave. Arch Bronconeumol (Engl Ed) 2021;57:717–9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Atkin C Kamwa V Reddy-Kolanu V, et al. : The changing characteristics of COVID-19 presentations. A regional comparison of SARS-CoV-2 hospitalised patients during the first and second wave. Acute Med 2021;20:92–100 [PubMed] [Google Scholar]
- 13.ATI : Role of inpatient rehabilitation hospitals during the COVID-19 pandemic 2021. Available at: https://atiadvisory.com/wp-content/uploads/2021/12/Role-of-Inpatient-Rehabilitation-Hospitals-During-the-COVID-19-Pandemic.pdf. Accessed December 22, 2021 [DOI] [PMC free article] [PubMed]
- 14.Medrinal C Prieur G Bonnevie T, et al. : Muscle weakness, functional capacities and recovery for COVID-19 ICU survivors. BMC Anesthesiol 2021;21:64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Tay MRJ Ong PL Puah SH, et al. : Acute functional outcomes in critically ill COVID-19 patients. Front Med (Lausanne) 2020;7:615997. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Jain E, Harmon EY, Sonagere MB: Functional outcomes and post-discharge care sought by patients with COVID-19 compared to matched controls after completing inpatient acute rehabilitation. PM R 2021;13:618–25 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hartsgrove C Guevarra-Fernandez J Kendall J, et al. : Measuring discharge outcomes, length of stay, and functional ADL score during Covid-19 in inpatient rehabilitation hospitals. Arch Phys Med Rehabil 2021;102:2291–9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Busching G Zhang Z Schmid JP, et al. : Effectiveness of pulmonary rehabilitation in severe and critically ill COVID-19 patients: a controlled study. Int J Environ Res Public Health 2021;18:8956. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.NIH : Clinical spectrum of SARS-CoV-2 infection. Updated April 21, 2021. Available at: https://www.covid19treatmentguidelines.nih.gov/overview/clinical-spectrum/. Accessed September 10, 2021
- 20.Final IRF-PAI Version 3.0 . Centers for Medicare and Medicaid Services. Updated August 1, 2019. 2021. Available at: https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/IRF-Quality-Reporting/IRF-PAI-and-IRF-PAI-Manual#:~:text=The%20IRF-PAI%20is%20the%20assessment%20instrument%20IRF%20providers,Medicare%20Part%20C%20patient%20discharged%20from%20an%20IRF. Accessed August 19, 2021 [PubMed]
- 21.RTI International: Specifications for the function quality measures adopted in the Inpatient Rehabilitation Facility Quality Reporting Program. Available at: https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/Post-Acute-Care-Quality-Initiatives/Downloads/IRF-QRP-Function-Quality-Measure-Specifications-August-2018.pdf. Accessed August 19, 2021
- 22.Chodosh J Edelen MO Buchanan JL, et al. : Nursing home assessment of cognitive impairment: development and testing of a brief instrument of mental status. J Am Geriatr Soc 2008;56:2069–75 [DOI] [PubMed] [Google Scholar]
- 23.Kaushasky KLM Prchal J Levi M, et al. : Williams Hematology. New York, NY, McGraw-Hill Education, 2016 [Google Scholar]
- 24.Groll DL To T Bombardier C, et al. : The development of a comorbidity index with physical function as the outcome. J Clin Epidemiol 2005;58:595–602 [DOI] [PubMed] [Google Scholar]
- 25.Kim HY: Statistical notes for clinical researchers: assessing normal distribution (2) using skewness and kurtosis. Restor Dent Endod 2013;38:52–4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Hermann M Pekacka-Egli AM Witassek F, et al. : Feasibility and efficacy of cardiopulmonary rehabilitation after COVID-19. Am J Phys Med Rehabil 2020;99:865–9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Paneroni M Simonelli C Saleri M, et al. : Muscle strength and physical performance in patients without previous disabilities recovering from COVID-19 pneumonia. Am J Phys Med Rehabil 2021;100:105–9 [DOI] [PubMed] [Google Scholar]
- 28.Olezene CS Hansen E Steere HK, et al. : Functional outcomes in the inpatient rehabilitation setting following severe COVID-19 infection. PLoS One 2021;16:e02e48824. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Petrilli CM Jones SA Yang J, et al. : Factors associated with hospital admission and critical illness among 5279 people with coronavirus disease 2019 in New York City: prospective cohort study. BMJ 2020;369:m1966. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Cuomo A: in SoN Y. (ed): State of New York Executive Order No. 202.30: Continuing Tempory Suspension and Modification of Laws Relating to the Disaster Emergency : 2020. Available at: https://www.governor.ny.gov/sites/default/files/atoms/files/EO202.30.pdf. Accessed August 25, 2021 [Google Scholar]
- 31.Herridge MS Cheung AM Tansey CM, et al. : One-year outcomes in survivors of the acute respiratory distress syndrome. N Engl J Med 2003;348:683–93 [DOI] [PubMed] [Google Scholar]
- 32.NIH : COVID-19 Treatment Guidelines Panel. Coronavirus disease 2019 (COVID-10) treatment guidelines. National Institutes of Health. NOH. Updated August 25, 2021. Available at: https://www.covid19treatmentguidelines.nih.gov/management/clinical-management/hospitalized-adults--therapeutic-management. Accessed September 10, 2021 [PubMed] [Google Scholar]
- 33.Dawson C Capewell R Ellis S, et al. : Dysphagia presentation and management following COVID-19: an acute care tertiary centre experience. J Laryngol Otol 2020:1–6. doi: 10.1017/S0022215120002443 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Neevel AJ Smith JD Morrison RJ, et al. : Postacute COVID-19 laryngeal injury and dysfunction. OTO Open 2021;5:2473974X211041040. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Munoz M Romero A Morales M, et al. : Iron metabolism, inflammation and anemia in critically ill patients. A cross-sectional study. Nutr Hosp 2005;20:115–20 [PubMed] [Google Scholar]
- 36.DeBellis RJ: Anemia in critical care patients: incidence, etiology, impact, management, and use of treatment guidelines and protocols. Am J Health Syst Pharm 2007;64(3 suppl 2):S14–21 [DOI] [PubMed] [Google Scholar]
- 37.Taherifard E Taherifard E Movahed H, et al. : Hematologic autoimmune disorders in the course of COVID-19: a systematic review of reported cases. Hematology 2021;26:225–39 [DOI] [PubMed] [Google Scholar]
- 38.Claflin ES Daunter AK Bowman A, et al. : Hospitalized patients with COVID-19 and neurological complications experience more frequent decline in functioning and greater rehabilitation needs. Am J Phys Med Rehabil 2021;100:725–9 [DOI] [PubMed] [Google Scholar]
- 39.Leigh AE McCall J Burke RV, et al. : Predictors of functional dependence after COVID-19: a retrospective examination among veterans. Am J Phys Med Rehabil 2021;100:34–8 [DOI] [PubMed] [Google Scholar]
- 40.Centers for Medicare & Medicaid Services (CMS): Medicare Program; Inpatient Rehabilitation Facility (IRF) Prospective Payment System for Federal Fiscal Year 2020 and Updates to the IRF Quality Reporting Program 39054-39067 (2019). Available at: https://www.govinfo.gov/content/pkg/FR-2019-08-08/pdf/2019-16603.pdf. Accessed August 25, 2021