Improving hospital outcomes using an acute hospital rehabilitation intensive service (ARISE) for patients with COVID-19

Dear Editor, The initial surge of COVID-19 in 2020 placed an unprecedented strain on the hospital healthcare system [1]. In addition to critical respiratory complications, there was early recognition that COVID-19 could lead to significant hospital-acquired deficits with mobility, swallowing, and performance of activities of daily living [2]. In usual care, these impairments would require patients to receive rehabilitation in a post-acute care rehabilitation facility prior to returning home. However, post-acute care discharges were more challenging due to quarantine requirements, and many facilities could not accommodate isolation precautions. This led to growing concerns from clinical and operational leaders that these factors could significantly impact capacity management due to longer lengths of hospital stay and patient needs after discharge. To address these challenges, the Johns Hopkins Physical Medicine and Rehabilitation department developed a novel program deployed in the acute care setting. The acute hospital rehabilitation intensive service (ARISE) program was built on existing evidence supporting early rehabilitation for critically ill patients and concepts of acute inpatient rehabilitation that focused on early, individualized, intensive, and coordinated interdisciplinary rehabilitation to maximize functional recovery [3, 4]. Here, we sought to evaluate the impact of the ARISE program on the rate of discharge to home and length of stay in patients with COVID-19. Adult patients admitted with a diagnosis of COVID-19 between 13 April and 30 June 2020 were included in this retrospective, observational study. Those admitted for ≥5 days and received ARISE program at The Johns Hopkins Hospital were compared to those who did not receive ARISE in three hospitals within the same health system. Patients were excluded from analysis if they died. This study was approved by The Johns Hopkins Medicine Institutional Review Board. Patients admitted with COVID-19 and medically stable to participate in rehabilitation were evaluated by a multidisciplinary team, including physical therapists (PT), occupational therapists (OT), speech language pathologists (SLP), and physiatrist. They assessed daily functional impairments using the AM-PAC 6-Clicks Inpatient Basic Mobility and Daily Activity short forms [5]. ARISE eligible patients had AM-PAC raw scores 13–21 (range: 6–24). The ARISE care model consists of delivering rehabilitation interventions [4]. This approach is in stark contrast to usual hospital rehabilitation care where therapists and physiatrists mostly play a consultative role providing discharge recommendations. The primary and secondary outcomes were discharge destination (home vs. nonhome) and length of stay, respectively. We used propensity score methodology to the nearest neighbor algorithm to match age, race, gender, initial sequential organ failure assessment (SOFA) score, body mass index, third-party payer, comorbidity count, area of deprivation index, prior admission setting, need for intensive care unit (ICU) stay, and ventilator in the first 2 days of admission. After matching, we evaluated odds of discharge home with logistic regression and length of stay with linear regression across the ARISE and non-ARISE hospitals (Stata teffects package; Stata Version 17.0, StataCorp, College Station, TX, USA). A total of 218 eligible patients were included in the analysis (ARISE: N = 132, non-ARISE: N = 86). Compared to the non-ARISE hospitals, patients in the ARISE hospital were younger, lived in more socioeconomically disadvantaged areas, were more often admitted from another acute facility, were more frequently admitted to the ICU or ventilated within 48 h, had higher admission SOFA scores, and had more days with delirium (Table 1). Coordinated visit frequency of PT, OT, and SLP had similar frequency per day in the ICU (0.42 vs. 0.32, p = 0.32) but was significantly higher in the non-ICU setting (0.54 vs. 0.35, p = 0.003). In propensity matched analysis, patients in the ARISE hospital were more likely to be discharged home versus non-ARISE (OR = 1.17, 95% CI, 1.04–1.32; p = 0.009). No differences in LOS were observed between the two groups (mean [SD], ARISE: 13.8 days [12.1]; non-ARISE: 11.4 [7.3] days, p = 0.214). As expected, AM-PAC mobility scores were higher for patients discharged home versus facility (mean [SD] home: 23[3], facility: 17[6]; p < 0.001). Patients admitted to the ARISE hospital with COVID-19, despite being sicker and from more socioeconomically disadvantaged areas, were more likely to be discharged home. Our findings suggest the importance of early involvement of a multidisciplinary rehabilitation team delivering coordinated rehabilitation at a higher frequency in this patient population. It also highlights the value of using a standardized measure (AM-PAC 6-Clicks) to select the right patient for appropriate rehabilitation interventions [6-8]. Of note, our study cannot exclude that financial differences played a role on discharge destination. The ARISE care delivery model is likely to also be impactful in other inpatient groups who experience acute disability. The authors have no conflict of interests to declare.