key: cord-0979227-m9ws5rmv
authors: Johnson, Joshua K; Lapin, Brittany; Green, Karen; Stilphen, Mary
title: Frequency of Physical Therapist Intervention Is Associated With Mobility Status and Disposition at Hospital Discharge for Patients With COVID-19
date: 2020-09-28
journal: Phys Ther
DOI: 10.1093/ptj/pzaa181
sha: 8f7014f7d16561bec2e74e72e9d2ace4ff59e410
doc_id: 979227
cord_uid: m9ws5rmv

OBJECTIVE: For patients diagnosed with the novel coronavirus, COVID-19, evidence is needed to understand the effect of treatment by physical therapists in the acute hospital on patient outcomes. The primary aims of this study were to examine the relationship of physical therapy visit frequency and duration in the hospital with patients’ mobility status at discharge and probability of discharging home. METHODS: This retrospective study included patients with COVID-19 admitted to any of eleven hospitals in one health system. The primary outcome was mobility status at discharge, measured using the Activity Measure for Post-Acute Care 6-Clicks basic mobility (6-Clicks mobility) and the Johns Hopkins Highest Level of Mobility (JH-HLM) scales. Discharge to home vs. to a facility was a secondary outcome. Associations between these outcomes and physical therapy visit frequency or mean duration were tested using multiple linear or modified Poisson regression. Potential moderation of these relationships by particular patient characteristics was examined using interaction terms in subsequent regression models. RESULTS: For the 312 patients included, increased physical therapy visit frequency was associated with higher 6-Clicks mobility (b = 3.63; 95% CI = 1.54–5.71) and JH-HLM scores (b = 1.15; 95% CI = 0.37–1.93) at hospital discharge and with increased probability of discharging home (adjusted relative risk = 1.82; 95% CI = 1.25–2.63). Longer mean visit duration was also associated with improved mobility at discharge and the probability of discharging home, though the effects were less pronounced. Few moderation effects were observed. CONCLUSION: Patients with COVID-19 demonstrated improved mobility at hospital discharge and higher probability of discharging home with increased frequency and longer mean duration of physical therapy visits. These associations were not generally moderated by patient characteristics. IMPACT: Physical therapy should be an integral component of care for patients hospitalized due to COVID-19. Providing sufficient physical therapist interventions to improve outcomes must be balanced against protection from viral spread.

The overall cumulative hospitalization rate associated with the novel coronavirus disease, COVID-19, is 120.9 per 100,000 population and continuously increasing. 1 COVID-19 is associated with varying levels of illness severity and a multitude of symptoms. 2 Those who are hospitalized, especially with critical illness, most often present with respiratory symptoms including acute respiratory distress syndrome. [3] [4] [5] In patients with respiratory disease not related to COVID-19, greater illness severity in the hospital is associated with greater deficits in mobility. 6 This loss of mobility has been shown in previous studies to be amenable to improvement via rehabilitation interventions. [7] [8] [9] Limited mobility at the time of hospital discharge is also associated with greater likelihood of discharge to post-acute care (PAC) facilities. [10] [11] [12] In the era of COVID-19, and especially for patients with active COVID-19 disease or those in high-risk groups, efforts to limit spread have caused health systems to focus on increasing the proportion of patients that are discharged from the hospital to home. 13, 14 Since physical therapists providing care in the acute hospital setting play a role in improving patients' mobility status prior to hospital discharge, 15 their interventions may positively influence patients' ability to discharge home from the hospital rather than to a PAC facility for ongoing rehabilitation.

Specifically for patients with COVID-19, evidence is needed to understand the effect of treatment by physical therapists in the hospital on mobility status at hospital discharge and the likelihood of discharging home. The primary aim of this study was to examine, for patients hospitalized with COVID- We also analyzed the relationship between physical therapy visit frequency and the likelihood of discharge to home as a secondary outcome. In secondary analyses, we examined the effect of the mean duration of individual physical therapy treatment visits on these same outcomes.

This was a retrospective study using records for patients admitted to one of eleven acute care hospitals in one health system. Patients were included if they had a confirmed positive test (via nasopharyngeal swab) for COVID-19 either during hospitalization or which resulted in a hospitalization, had been discharged from the hospital by June 10, 2020, and had been evaluated by a physical therapist during their hospital stay. Patients were excluded from the analyses of mobility status at discharge if only one score was recorded from the 6-Clicks mobility or JH-HLM. Patients who died in the hospital or were discharged with hospice services (since their prognosis for recovery was poor) were excluded when analyzing the likelihood of discharging to home.

Data were extracted from the Cleveland Clinic (CCHS) COVID-19 research data registry and the Rehabilitation and Related Outcomes Learning Lab, a data registry that links patient demographic and rehabilitation care episode data across CCHS-affiliated care settings. The CCHS IRB approved this study and provided a waiver of informed consent since these data were accessed retrospectively.

For the primary analysis, the predictor variable was the frequency of physical therapy visits. This was calculated as the total number of completed visits divided by the number of days from the date of the physical therapist evaluation through the date of hospital discharge. In secondary analyses, we included

the mean duration of individual physical therapy treatment visits as the primary predictor variable. This was calculated as the mean of billed minutes of physical therapist interventions for all completed visits.

To better understand care utilization patterns, we also described the number of days from hospital admission to physical therapist evaluation, the number of days from physical therapist evaluation to hospital discharge, the count of completed physical therapy visits, and the count of attempted physical therapy visits (including completed visits).

Mobility status at the time of hospital discharge was the primary outcome and was evaluated separately using scores from the 6-Clicks mobility and the JH-HLM. The 6-Clicks mobility is a valid and reliable measure of mobility in the acute hospital setting. 16, 17 It assesses a patient's capability to complete basic transferring and ambulation tasks. Higher 6-Clicks mobility scores (which range from 6 to 24) indicate greater levels of independence with those tasks. The JH-HLM captures the level of activity that patients complete within a specific timeframe. JH-HLM scores range from 1 (only lying) to 8 (walking 250+ feet). 18 In CCHS hospitals, both the 6-Clicks mobility and the JH-HLM are scored by physical therapists for each completed visit.

The likelihood of discharge to home versus to a PAC facility was analyzed as a secondary outcome. Patients were considered to have discharged to home regardless of whether or not home health services were in place.

Patient demographic and clinical characteristics were described for the full cohort of patients evaluated by a physical therapist during their hospital admission. We also described these characteristics for the samples included for the analysis of each outcome and compared them to the full cohort to evaluate their representativeness.

We examined the adjusted effect of physical therapy visit frequency and mean visit duration on patients' mobility status at discharge using multiple linear regression. To examine the adjusted effect of physical therapy visit frequency and mean visit duration on patients' likelihood of discharging to home vs. to a PAC facility, we estimated adjusted relative risk (aRR) using a modified Poisson regression with robust variance estimation. 19 The mean visit duration for each patient was divided by 10 for use in the models where it was the primary predictor. In all models, we controlled for patient-level covariates that are associated with other important outcomes including hospital readmission, 20-23 hospital discharge disposition, 11, 24 and morbidity or mortality. 25, 26 These included patient demographics (age, sex, and race), the primary payer for the episode of care, the hospital in which the patient was treated, and clinical factors including medical complexity as indicated by the All Patient Refined (APR) Diagnostic Related Group

Illness Severity modifier-a valid 27, 28 and reliable predictor 29 of in-hospital mortality across clinical populations-and whether the patient was admitted to the intensive care unit (ICU) during their hospitalization. Overall hospital length of stay (LOS) was included as a covariate in the models where the mean visit duration was the primary predictor variable, but excluded where physical therapy visit frequency was the primary predictor since our calculation of frequency accounted for a portion of the LOS. For the analysis of 6-Clicks mobility score at hospital discharge, we included the first recorded 6-Clicks mobility score as a covariate and similarly included the first recorded JH-HLM score as a covariate when analyzing the JH-HLM score at hospital discharge.

To aid interpretation, marginal estimation from the multivariable models was used to obtain predicted 6-Clicks mobility scores and JH-HLM scores at discharge and probabilities of discharge to home at representative values of physical therapy visit frequency and mean visit duration.

As an exploratory analysis to understand whether the relationships between physical therapy visit frequency or mean visit duration and mobility status at discharge or discharge to home were moderated by patient characteristics, we included interaction terms in separate regression models. We evaluated for moderation effects between physical therapy visit frequency or mean visit duration and initial 6-Clicks

mobility (or JH-HLM) score, age, sex, race, APR severity, or an ICU admission. We considered that a moderating effect may be present if the interaction term in the respective model was statistically significant (α<.05).

All statistical analyses were completed using Stata version 15.1 (College Station, Texas, USA) at a significance level of p<0.05. Sample size guidelines for linear and logistic regression indicate a minimum of 2-10 cases per independent variable. 30, 31 Given a maximum of 17 parameters in the primary analysis models, our minimum model sample size of 221 will achieve adequate power for our analyses.

As the results of this study are hypothesis-generating and focused on estimates of effects, there was no formal adjustment for multiple comparisons.

We identified 963 patients discharged from a CCHS hospital after treatment for COVID-19. The full study cohort included 312 patients who were evaluated by a physical therapist during their hospitalization. Their demographic and clinical characteristics are provided in Table 1 . Notably, the mean (SD) initial 6-Clicks mobility score was 14.7 (5.3) and initial JH-HLM score was 4.8 (1.7). Table 2 , the number of completed physical therapy visits was highly variable with a median of 3 and interquartile range (IQR) of 1 to 5 visits over, on average, 6 days between the date of the physical therapist evaluation and hospital discharge. The frequency of physical therapy visits was also variable with a mean of 0.5 visits per day (ie, one visit every other day), but which ranged from 0.1 to 1.5.

The mean visit duration was 25.3 minutes.

Of the 312 in the full cohort, 89 and 91 patients had only one 6-Clicks mobility and JH-HLM score recorded, respectively. The 89 patients with only one score recorded had only one documented visit during their hospitalization; a description of these patients' characteristics and discharge disposition can be found in Supplemental Table 1 .

The sample of 223 patients for whom 6-Clicks mobility scores at discharge were evaluated differed slightly from the full cohort in the proportion with an ICU stay (60.5% vs. 54.8%), overall hospital LOS (15 vs. 13 days), and first recorded 6-Clicks mobility score (13.9 vs. 14.7). After excluding those with only one 6-Clicks mobility score recorded, the median number of days from physical therapist evaluation to hospital discharge increased from 6 to 8 days and the median number of completed visits 

As shown in Table 3 , the mean (SD) 6-Clicks mobility score at discharge was 16 There were significant independent associations between visit frequency and both the 6-Clicks mobility 

There were 132 (46.8%) patients discharged home vs. to a PAC facility. Supplemental Table 2 shows the characteristics of these patients by discharge destination. In adjusted models, there was a statistically significant increase in likelihood of discharge to home with higher visit frequency (aRR=1.82; 95% CI = 1.25-2.63) and mean visit duration (1.22; 95% CI = 1.09-1.37). As shown in Table   4 , the probability of discharge to home increased with marginal increases in visit frequency. For example, the probability of discharge to home for patients with a visit frequency of 0.5 was 45% compared to a probability of 60% for patients with a visit frequency of 1.0.

We did not observe any statistically significant interaction effects between visit frequency and the patient characteristics we tested as potential moderators in the relationship between visit frequency and the outcomes of interest. There were a number of significant interaction effects using mean duration of observed for patient characteristics we tested as potential moderators in the relationship between mean visit frequency and the outcomes of interest.

In this examination of data for patients treated in one of eleven hospitals in a single health system for COVID-19, we identified that higher frequency of physical therapy visits and longer individual visits are both significantly associated with better mobility status at hospital discharge and with increased

10 probability of discharging to home. Further, with a few noted exceptions, our results suggest that individual patient characteristics do not moderate these observed associations. Thus, any patient with COVID-19 being treated by a physical therapist should have visits at a higher frequency and for longer durations than may be typical for most patients in the acute hospital setting in order to achieve higher 6-Clicks mobility and JH-HLM scores at discharge and to have a higher probability of discharging to home.

These findings are novel given our examination in patients with COVID-19, a novel disease. The few moderating effects that we observed in our exploratory analyses were between the mean duration of each visit-not the frequency of visits-and particular patient characteristics. Females were more likely than otherwise similar males to benefit from longer visits with better mobility at discharge and higher probability of discharging home. Longer visits did not improve the probability of discharge to home for patients who were in the ICU. These observations deserve further exploration in subsequent studies as more data become available. For the latter, it may be that physical therapists dedicated significant time providing treatment for patients with the most critical illness, but that these patients remained unable to discharge home primarily due to their medical status. Our sample size was inadequate to determine whether this is the case. particularly those who may develop post-intensive care syndrome. [52] [53] [54] In the acute care hospital, especially given our findings that patients benefit from more frequent and longer physical therapy visits, it is imperative that patients' needs for intervention be balanced with staff safety. This requires that protocols be established to maximize both. In CCHS hospitals, protocols have been developed in a collaborative effort between physicians and both frontline staff and management teams in physical therapy, occupational therapy, respiratory therapy, and nursing. At our main campus hospital and several regional hospitals, a COVID-19 team of therapists has been designated.

This team provides the preponderance of physical therapy care for patients with COVID-19. Daily 13 mobility sustainment when skilled therapy is not necessary. All of these tools and practices have been shared as "best practice" with all hospital therapy managers throughout CCHS. Their daily use is assured at the main campus and several regional hospitals.

Given its observational design, potential biases do exist in this study. Our models included several patient characteristics and clinical variables that are associated with the outcomes of interest, but our adjustment was not comprehensive. For example, more traditional markers of illness severity in the hospital (eg, ICU LOS or hours on mechanical ventilation) were missing for a large proportion of the sample (49.0% and 72.8%, respectively). As we noted, however, the APR severity modifier does have good evidence as a robust substitute. Unknown patient-level confounders, and/or their moderating relationships, could contribute to improved mobility at discharge and discharge to home.

Improved mobility at discharge was assessed only for patients who had at least two physical therapy visits. We are unable to determine from our data why some patients did not receive a physical therapist consult at all or were seen only once. The patient-level data do suggest that patients who were seen only once tended to have more mild illness as indicated by shorter LOS and higher initial 6-Clicks mobility and JH-HLM scores. Our sample size was not adequate to determine whether these observations were consistent in some measurable way, but this should be explored further with larger patient cohorts.

Further, the sample was drawn from only one health system so generalizability may be limited.

Additionally, the fact that there is an emphasis in hospitals to discharge patients home in order to limit spread of COVID-19 in PAC facilities could introduce some confounding by indication. That is, knowing this emphasis exists may predispose clinical decision-making on the part of physical therapists to both increase the frequency and duration of their visits and recommend that patients be discharged to home despite persistent functional deficits that would otherwise prompt a recommendation for additional rehabilitation in a PAC facility. We are unable to determine whether this was the case for patients in this

study, but the fact that >50% of patients did discharge to a PAC facility suggests that there may have been some clinical rationale for continuing to recommend PAC for patients with the need for such ongoing care.

Lastly, our study is limited to evaluating physical therapist intervention in terms of temporal factors (frequency and duration) rather than content. We assume that longer and more frequent visits enabled higher intensity interventions focused on activity, mobility, and strength training that would be meaningful for patient outcomes. We do not, however, have data available to confirm this assumption.

For patients with COVID-19, more frequent and longer physical therapy visits in the hospital are directly related to better mobility at discharge and greater probability for discharge to home. Physical therapist teams should collaborate with interprofessional colleagues to enable an adequate volume of physical therapist interventions to improve patient outcomes while ensuring safety from viral spread. As the COVID-19 pandemic continues to evolve, additional research will be needed to better understand and facilitate the role of acute care physical therapists in setting an optimal trajectory of functional recovery early in the course of patients' care.

Concept 

The Cleveland Clinic Institutional Review Board approved this study and provided a waiver of informed consent, as these data were accessed retrospectively.

There are no funders to report for this study.

The authors completed the ICMJE Form for Disclosure of Potential Conflicts of Interest and reported no conflicts of interest. 30 16.6 (16.1-17.1) 5.5 (5.3-5.7) 52% (46%-57%) 35 17.3 (16.7-18.0) 5.8 (5.5-6.0) 57% (49%-66%) 40 18.1 (17.2-19.1) 6.0 (5.7-6.4) 64% (52%-75%) 45 18.9 (17.6-20.2) 6.3 (5.8-6.8) 70% (54%-87%) a JH-HLM = Johns Hopkins Highest Level of Mobility.

a Calculated as the number of completed visits divided by the number of days from evaluation to hospital discharge

COVIDView: A Weekly Surveillance Summary of U.S. COVID-19 Activity | CDC

Clinical characteristics of coronavirus disease 2019 in China

Critical Care Utilization for the COVID-19 Outbreak in Lombardy, Italy: Early Experience and Forecast during an Emergency Response

Epidemiology, clinical course, and outcomes of critically ill adults with COVID-19 in New York City: a prospective cohort study

Characteristics and Outcomes of 21 Critically Ill Patients with COVID-19 in Washington State

Physical Function and Disability After Acute Care and Critical Illness Hospitalizations in a Prospective Cohort of Older Adults

Rehabilitation therapy and outcomes in acute respiratory failure: An observational pilot project

Early physical medicine and rehabilitation for patients with acute respiratory failure: a quality improvement project

The effects of active mobilisation and rehabilitation in ICU on mortality and function: a systematic review

AM-PAC "6-Clicks" functional assessment scores predict acute care hospital discharge destination

Routine Inpatient Mobility Assessment and Hospital Discharge Planning

Using a Decision Support Algorithm for Referrals to Post-Acute Care

We Should Move High-Intensity Postacute Care Home | Health Affairs

Postacute Care Preparedness for COVID-19

Physical Therapy in the COVID-19 Pandemic: Forging a Paradigm Shift for Rehabilitation in Acute Care

Validity of the AM-PAC "6-Clicks" inpatient daily activity and basic mobility short forms

Interrater Reliability of AM-PAC "6-Clicks" Basic Mobility and Daily Activity Short Forms

Promoting mobility and reducing length of stay in hospitalized general medicine patients: A quality-improvement project

A Modified Poisson Regression Approach to Prospective Studies with Binary Data

Potentially avoidable 30-day hospital readmissions in medical patients: derivation and validation of a prediction model

Association of impaired functional status at hospital discharge and subsequent rehospitalization

Risk prediction models for hospital readmission -A systematic review

Derivation and validation of an index to predict early death or unplanned readmission after discharge from hospital to the community

Standardized discharge planning assessments: impact on patient outcomes

Association of Loss of Independence With Readmission and Death After Discharge in Older Patients After Surgical Procedures

Slower walking speed forecasts increased postoperative morbidity and 1-year mortality across surgical specialties

Validation of the All Patient Refined Diagnosis Related Group (APR-DRG) Risk of Mortality and Severity of Illness Modifiers as a Measure of Perioperative Risk

Risk-adjusting acute myocardial infarction mortality: are APR-DRGs the right tool?

Comparison of Measures to Predict Mortality and Length of Stay in Hospitalized Patients

The number of subjects per variable required in linear regression analyses

Relaxing the rule of ten events per variable in logistic and cox regression

Additional Physical Therapy Services Reduce Length of Stay and Improve Health Outcomes in People With Acute and Subacute Conditions: An Updated Systematic Review and Meta-Analysis

An analysis of the relationship between the utilization of physical therapy services and outcomes of care for patients after total hip arthroplasty

Association of Rehabilitation Intensity for Stroke and Risk of Hospital Readmission

Use of Hospital-Based Rehabilitation Services and Hospital Readmission Following Ischemic Stroke in the United States

Functional recovery in the elderly after major surgery: Assessment of mobility recovery using wireless technology

Increasing Mobility via In-hospital Ambulation Protocol Delivered by Mobility Technicians: A Pilot Randomized Controlled Trial

Increasing patient mobility through an individualized goalcentered hospital mobility program: A quasi-experimental quality improvement project

Physical Therapists' Management of Patients in the Acute Care Setting: An Observational Study

Increased hospital-based physical rehabilitation and information provision after intensive care unit discharge: The RECOVER randomized clinical trial

Interpretation of changes in health-related quality of life: the remarkable universality of half a standard deviation

Interpreting rehabilitation outcome measurements

Recommendations for respiratory rehabilitation in adults with COVID-19

The role of physical and rehabilitation medicine in the COVID-19 pandemic: The clinician's view

Rehabilitation and respiratory management in the acute and early post-acute phase

Recommendations for Hospital-Based Physical Therapists Managing Patients With COVID-19. Phys Ther

Italian Physical Therapists' Response to the Novel COVID-19 Emergency

The place of early rehabilitation in ICU for Covid-19

COVID-19 Guide for the Rehabilitation Clinician

Rehabilitation following critical illness in people with COVID-19 infection

Prone Positioning of Patients With Acute Respiratory Distress Syndrome Related to COVID-19: A Rehabilitation-Based Prone Team. Phys Ther

The COVID-19 rehabilitation pandemic

The Essential Role of Home-and Community-Based Physical Therapists During the COVID-19 Pandemic

Home and Community-Based Physical Therapist Management of Adults With Post-Intensive Care Syndrome. Phys Ther