Skip to main content
Download PDF

Early physical rehabilitation dosage in the intensive care unit associates with hospital outcomes after critical COVID-19

Critical Care volume 28, Article number: 248 (2024) Cite this article

Abstract

Objective

To examine the relationship between physical rehabilitation parameters including an approach to quantifying dosage with hospital outcomes for patients with critical COVID-19.

Design

Retrospective practice analysis from March 5, 2020, to April 15, 2021.

Setting

Intensive care units (ICU) at four medical institutions.

Patients

n = 3780 adults with ICU admission and diagnosis of COVID-19.

Interventions

We measured the physical rehabilitation treatment delivered in ICU and patient outcomes: (1) mortality; (2) discharge disposition; and (3) physical function at hospital discharge measured by the Activity Measure-Post Acute Care (AM-PAC) “6-Clicks” (6–24, 24 = greater functional independence). Physical rehabilitation dosage was defined as the average mobility level scores in the first three sessions (a surrogate measure of intensity) multiplied by the rehabilitation frequency (PT + OT frequency in hospital).

Measurements and main results

The cohort was a mean 64 ± 16 years old, 41% female, mean BMI of 32 ± 9 kg/m2 and 46% (n = 1739) required mechanical ventilation. For 2191 patients who received rehabilitation, the dosage and AM-PAC at discharge were moderately, positively associated (Spearman’s rho [r] = 0.484, p < 0.001). Multivariate linear regression (model adjusted R2 = 0.68, p < 0.001) demonstrates mechanical ventilation (β = − 0.86, p = 0.001), average mobility score in first three sessions (β = 2.6, p < 0.001) and physical rehabilitation dosage (β = 0.22, p = 0.001) were predictive of AM-PAC scores at discharge when controlling for age, sex, BMI, and ICU LOS.

Conclusions

Greater physical rehabilitation exposure early in the ICU is associated with better physical function at hospital discharge.

Brief report

Objectives

Exercise and early mobility are key components of clinical practice guidelines for patients with critical illness, as defined in the Intensive Care Unit (ICU) Liberation bundle [1]. However, findings from multiple randomized ICU rehabilitation trials have been equivocal, demonstrating minimal impact on mortality and physical function [2,3,4,5]. A potential explanation for the lack of benefits is a non-specific exercise dose. Patients are routinely randomized to “one-size-fits-all” protocols leading to heterogeneity in the response to treatment. Dosage that accounts for the frequency and intensity of exercise is frequently overlooked or not addressed in critical care practice and research. The rehabilitation dosage delivered in large randomized controlled trials (RCTs) is rarely implemented in clinical practice [6, 7], and patients seldomly receive a targeted or individualized dose of exercise. Patients with critical COVID-19 have not been studied to determine if dosage of exercise is related to outcomes. The COVID-19 pandemic may have unintentionally altered patterns in rehabilitation practice due to periods of isolation [8]. Thus, the primary objective of this study was to examine the relationship between physical rehabilitation parameters including an approach to quantifying dosage with hospital outcomes for patients with critical COVID-19.

Design

Retrospective practice analysis for patients hospitalized from March 5, 2020, to April 15, 2021.

Setting

ICUs at four academic medical institutions (University of Kentucky, Cleveland Clinic, University of Michigan, and University of Southern California).

Patients

3780 adults (≥ 18 years of age) admitted to ICU with primary diagnosis of COVID-19.

Interventions

We examined the relationship between ICU-based physical rehabilitation interventions and hospital-based outcomes. Outcomes included: (1) mortality; (2) discharge disposition; and 3) physical function at or near hospital discharge measured by the Activity Measure-Post Acute Care (AM-PAC) “6-Clicks” Inpatient Mobility Short Form (6–24, 24 = greater functional independence) [9]. Physical rehabilitation parameters included time to first rehabilitation (physical [PT] or occupational [OT]) session in days, number of PT and OT sessions completed during hospital length of stay (LOS), frequency of PT and OT (# of session/hospital LOS), mobility status during first three and the last recorded (if more than 3 sessions) rehabilitation sessions. Mobility levels were quantified by the John Hopkins-Highest Level of Mobility (JH-HLM, 1–8, 1 = lying in bed; 8 = ambulating > 250 feet). The physical rehabilitation dose was quantified as the average JH-HLM score over the first three sessions (a surrogate measure of early intensity) multiplied by the rehabilitation frequency (PT + OT frequency). The dose provides information on delivery of ICU rehabilitation such that patients who achieve high mobility with daily frequency of rehabilitation receive the highest dosage, whereas patients with lower mobility levels and infrequent rehabilitation receive the lowest dose. Our method is based on our previous published studies demonstrating that the mobility levels obtained in the first 3 rehabilitation sessions predict, or at minimum, associate with patient-centered outcomes [10, 11].

Measurements and main results

Descriptive statistics were reported as mean ± SD, median [IQR], or n (%) as appropriate. A total of 3780 patients with COVID critical illness were included. Patients were stratified into groups according to discharge disposition (in-hospital death, subacute or long-term care facility, acute rehabilitation facility, home with services, or home independent). The change in mobility level during rehabilitation as measured by JH-HLM among discharge groups were compared using a two-way ANOVA. Dose of rehabilitation between discharge disposition groups were compared using Tukey’s multiple comparison test. Univariate analyses (Spearman’s correlation) were performed to assess associations between rehabilitation parameters and functional outcomes. Multivariate linear regression was performed to analyze the association between rehabilitation dose and discharge AM-PAC scores, which defined physical function among survivors, adjusting for pre-specified covariates including age, sex, body mass index (BMI), ICU length of stay, and receipt of mechanical ventilation.

Patient demographics are described in Table 1. The cohort was a mean 64 ± 16 years old, 41% female and mean BMI of 32 ± 9 kg/m2. Mechanical ventilation was required in 46% (n = 1739), and the median hospital LOS was 12 days (IQR 7–21). A total of 2200 (58%) and 1698 (45%) patients received at least one PT and OT session, respectively. The first rehabilitation session occurred 7.5 ± 8.0 days after ICU admission. Patients received PT at a frequency of 0.22 ± 0.14 days a week and OT at a frequency of 0.18 ± 0.11 days a week, equivalent to 2.8 rehabilitation sessions per week. Mobility levels on the JH-HLM scale generally increased from the first to last session (+ 0.93 ± 2.1). The mean JH-HLM score for all sessions was 4.6 ± 1.7; this suggests a likely ability to transfer from a bed to a chair but not stand for up to one minute. The mean dose of physical rehabilitation was 1.8 ± 1.3 units.

Table 1 Demographic, clinical and rehabilitation parameters
Full size table

Patients who died in the hospital (n = 994, 26%) were older, more likely to require mechanical ventilation, had longer durations of mechanical ventilation, less likely to receive PT or OT, and had longer ICU LOS (Table 1) compared to patients who survived to hospital discharge. Compared to survivors, those who died in the hospital had an earlier start of rehabilitation, but had lower frequencies of rehabilitation, achieved lower levels of mobility, and received a lower dose of physical rehabilitation (Table 1). Stratified by discharge disposition, patients discharged to home had the highest dose of rehabilitation (F = 69, p < 0.0001; Fig. 1).

Fig. 1
figure 1

Dosage and change in mobility during physical rehabilitation grouped based on discharge disposition. A The change in mobility level measured by John Hopkins-Highest Level of Mobility (JH-HLM, 0–8) stratified by discharge disposition (black—mortality in hospital; red = secondary facilty; purple = acute rehabilitation facilty; blue = home with services; green = home without services). Two-Way ANOVA demonstrated significant difference in JH-HLM scores based on group (F = 240.8, p < 0.0001) and change over rehabilitation sessions (F = 11.13, p < 0.00) with interaction (F = 6.7, p < 0.0001). B The dosage of rehabilitation (JH-HLM average over four reported sessions multiplied by the frequency of rehabilitation) is significantly different based on discharge disposition (F = 69, p < 0.0001) with Tukey’s multiple comparison test revealing significant differences at the p < 0.0001 denoted with ****; p < 0.001 denoted with **, and * denoting p < 0.05. C (Table) provides raw data for the mobility sessions during the first 3 (1–3) sessions and the last rehabilitation (4) stratified based on the discharge destination

Full size image

For 2191 patients who received PT or OT treatment and functional data were available in the EMR, the mean AM-PAC scores at discharge were 15.6 ± 5.9; suggesting a high-level of assistance is needed for bed-to-chair transfers [12]. Rehabilitation dose and physical function as measured by AM-PAC at discharge were moderately, positively associated (Spearman’s rho [r] = 0.484, p < 0.001). AM-PAC at discharge was also significantly associated with average mobility achieved in first 3 sessions (r = 0.799, p < 0.001), change in mobility from first to last session (r = 0.445, p < 0.001), and PT and OT frequency with physical function (r = 0.130, p < 0.001). Multivariate linear regression (model adjusted R2 = 0.68, p < 0.001) demonstrates the receipt of mechanical ventilation (β = − 0.86, p = 0.001), average mobility score in first three sessions (β = 2.6, p < 0.001) and physical rehabilitation dosage (β = 0.22, p = 0.001) were predictive of AM-PAC scores at discharge when controlling for age, sex, BMI, and ICU LOS.

Conclusions

Our study, involving over 2000 critically ill patients with COVID-19 at four academic medical centers, underscores the pivotal role of physical rehabilitation exposure in the ICU, demonstrating a significant correlation with favorable hospital outcomes. These findings align with previous research highlighting the positive association between rehabilitation frequency and outcomes in COVID-19 patients, albeit not specifically focusing on those with critical illness [13]. We measured rehabilitation dose by assessing both session frequency and a surrogate marker of intensity derived from achieved mobility levels during sessions. Although our approach lacks physiologic dosage markers such as vital signs, timing of rehabilitation, and duration of intensity, it introduces a method for quantifying dose with a single unit. Adopting this approach may enable ICU rehabilitation programs to specifically delineate the intervention provided and anticipate patient benefit. In future, stratification and phenotypic analysis accounting for dose hold promise to guide interventions in clinical research settings.

Precision exercise, dose–response, and individual response heterogeneity are concepts in exercise dosing that are known to impact outcomes in older adults and diverse patient populations [14,15,16]. The intra-individual variations have also been recognized in critical care with ventilatory and pharmaceutical interventions [17, 18]. To improve the ICU rehabilitation field, it is imperative that clinicians and researchers examine the dose–response. As clinicians, we must modify and adjust treatments based on patient- and clinical-related factors to enhance outcomes. Our work as well as other research demonstrates that older individuals and patients with chronic disease respond differently to rehabilitation and exercise interventions [11, 19]. Thus, we strongly suggest that researchers and clinicians must begin to examine the response to targeted or individualized rehabilitation dose. The approach using mobility as a surrogate marker of intensity has limitation and may be further strengthen by including physiological response and timing.

Our study has several strengths including a large sample size from multiple academic medical centers with functioning ICU rehabilitation programs. We also used real world data directly from each rehabilitation session provided to create our method. It is important to highlight that a large percentage of patients did not receive or received a very low frequency of PT and OT in the ICU. Low delivery of mobilization and rehabilitation in the ICU remains common in clinical practice in ICU [6, 20] even despite clinical practice guidelines [1]. Disparate delivery of rehabilitation in practice as well as missing or limited functional mobility data in the EMR may have limited our findings. The retrospective design limits our ability to draw definitive conclusions with regard to causation and is at risk of residual confounding, so our findings should be considered hypothesis-generating. The study design may introduce representation and selection biases as it is impossible to control for all potential confounders, although, our modeling did account for receipt of MV and ICU length of stay. Still yet, patients who survived ICU may have had lower severity of illness and participated at higher levels of mobility with rehabilitation, regardless of the dose. Lastly, we could not account for duration of exercise intensity in our models; longer physiologic demands may produce additional benefits which warrants future investigations. In conclusion, we found that greater dose of rehabilitation during critical illness due to COVID-19 was associated with improved outcomes. Future studies should utilize personalized rehabilitation doses and identify the most optimal personalized dosage of rehabilitation in critically ill patients, including those with COVID-19.

Availability of data and materials

The datasets generated and/or analyzed during the current study are not publicly available due ongoing analyses and secondary studies by the authors, but are available from the corresponding author on reasonable request.

References

  1. Devlin JW, Skrobik Y, Gélinas C, et al. Clinical practice guidelines for the prevention and management of pain, agitation/sedation, delirium, immobility, and sleep disruption in adult patients in the ICU. Crit Care Med. 2018;46(9):e825–73. https://doi.org/10.1097/ccm.0000000000003299. (in Eng).

    Article  PubMed  Google Scholar 

  2. Morris PE, Berry MJ, Files DC, et al. Standardized rehabilitation and hospital length of stay among patients with acute respiratory failure: a randomized clinical trial. JAMA. 2016;315(24):2694–702. https://doi.org/10.1001/jama.2016.7201.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  3. Moss M, Nordon-Craft A, Malone D, et al. A randomized trial of an intensive physical therapy program for patients with acute respiratory failure. Am J Respir Crit Care Med. 2016;193(10):1101–10. https://doi.org/10.1164/rccm.201505-1039OC. (in Eng).

    Article  PubMed  PubMed Central  Google Scholar 

  4. Tipping CJ, Harrold M, Holland A, Romero L, Nisbet T, Hodgson CL. The effects of active mobilisation and rehabilitation in ICU on mortality and function: a systematic review. Intensive Care Med. 2017;43(2):171–83. https://doi.org/10.1007/s00134-016-4612-0. (in Eng).

    Article  PubMed  Google Scholar 

  5. Hodgson CL, Bailey M, Bellomo R, et al. Early active mobilization during mechanical ventilation in the ICU. N Engl J Med. 2022;387(19):1747–58. https://doi.org/10.1056/NEJMoa2209083. (in Eng).

    Article  PubMed  Google Scholar 

  6. Jolley SE, Moss M, Needham DM, et al. Point prevalence study of mobilization practices for acute respiratory failure patients in the United States. Crit Care Med. 2017;45(2):205–15. https://doi.org/10.1097/ccm.0000000000002058. (in Eng).

    Article  PubMed  PubMed Central  Google Scholar 

  7. Nydahl P, Ruhl AP, Bartoszek G, et al. Early mobilization of mechanically ventilated patients: a 1-day point-prevalence study in Germany. Crit Care Med. 2014;42(5):1178–86. https://doi.org/10.1097/ccm.0000000000000149. (in Eng).

    Article  PubMed  Google Scholar 

  8. Myszenski A, Bello R, Melican C, Pfitzenmaier N. Patient characteristics and acute PT and OT utilization during the initial surge of COVID-19: a retrospective observational study. J Acute Care Phys Ther. 2022;13(1):2–7. https://doi.org/10.1097/jat.0000000000000163. (in Eng).

    Article  PubMed  Google Scholar 

  9. Jette DU, Stilphen M, Ranganathan VK, Passek SD, Frost FS, Jette AM. Validity of the AM-PAC “6-Clicks” inpatient daily activity and basic mobility short forms. Phys Ther. 2014;94(3):379–91. https://doi.org/10.2522/ptj.20130199. (in Eng).

    Article  PubMed  Google Scholar 

  10. Mayer KP, Pastva AM, Du G, et al. Mobility levels with physical rehabilitation delivered during and after extracorporeal membrane oxygenation: a marker of illness severity or an indication of recovery? Phys Ther. 2022. https://doi.org/10.1093/ptj/pzab301. (in Eng).

    Article  PubMed  Google Scholar 

  11. Mayer KP, Silva S, Beaty A, et al. Relationship of age and mobility levels during physical rehabilitation with clinical outcomes in critical illness. Arch Rehabil Res Clin Transl. 2023;5(4):100305. https://doi.org/10.1016/j.arrct.2023.100305. (in Eng).

    Article  PubMed  PubMed Central  Google Scholar 

  12. Jette AM, Tao W, Norweg A, Haley S. Interpreting rehabilitation outcome measurements. J Rehabil Med. 2007;39(8):585–90. https://doi.org/10.2340/16501977-0119. (in Eng).

    Article  PubMed  Google Scholar 

  13. Johnson JK, Lapin B, Green K, Stilphen M. Frequency of physical therapist intervention is associated with mobility status and disposition at hospital discharge for patients with COVID-19. Phys Ther. 2021;101(1):pzaa181. https://doi.org/10.1093/ptj/pzaa181.

    Article  PubMed  Google Scholar 

  14. Ross R, Goodpaster BH, Koch LG, et al. Precision exercise medicine: understanding exercise response variability. Br J Sports Med. 2019;53(18):1141–53. https://doi.org/10.1136/bjsports-2018-100328. (in Eng).

    Article  PubMed  Google Scholar 

  15. Sanders LMJ, Hortobágyi T, la Bastide-van GS, van der Zee EA, van Heuvelen MJG. Dose-response relationship between exercise and cognitive function in older adults with and without cognitive impairment: a systematic review and meta-analysis. PLoS ONE. 2019;14(1):e0210036. https://doi.org/10.1371/journal.pone.0210036. (in Eng).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Herold F, Törpel A, Hamacher D, et al. Causes and consequences of interindividual response variability: a call to apply a more rigorous research design in acute exercise-cognition studies. Front Physiol. 2021;12:682891. https://doi.org/10.3389/fphys.2021.682891. (in Eng).

    Article  PubMed  PubMed Central  Google Scholar 

  17. Buckel M, Maclean P, Knight JC, Lawler PR, Proudfoot AG. Extending the ‘host response’ paradigm from sepsis to cardiogenic shock: evidence, limitations and opportunities. Crit Care. 2023;27(1):460. https://doi.org/10.1186/s13054-023-04752-8.

    Article  PubMed  PubMed Central  Google Scholar 

  18. Maslove DM, Tang B, Shankar-Hari M, et al. Redefining critical illness. Nat Med. 2022;28(6):1141–8. https://doi.org/10.1038/s41591-022-01843-x.

    Article  CAS  PubMed  Google Scholar 

  19. Jones JRA, Karahalios A, Puthucheary ZA, et al. Responsiveness of critically ill adults with multimorbidity to rehabilitation interventions: a patient-level meta-analysis using individual pooled data from four randomized trials*. Crit Care Med. 2023;51(10):1373–85. https://doi.org/10.1097/ccm.0000000000005936.

    Article  PubMed  Google Scholar 

  20. Prohaska CC, Sottile PD, Nordon-Craft A, et al. Patterns of utilization and effects of hospital-specific factors on physical, occupational, and speech therapy for critically ill patients with acute respiratory failure in the USA: results of a 5-year sample. Crit Care. 2019;23(1):175. https://doi.org/10.1186/s13054-019-2467-9. (in Eng).

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

The authors would like to acknowledge Joshua Veith, MD; Karna Sarin, MD; Eileen Bishop, DO; Heather Torbic, PharmD; Kathryn Bash, APRN, CNP; Maria Holztrager, APRN, PA; Jill O'Brien, PA-C.

Funding

Dr. Kirby Mayer was supported by the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institute of Health K23-AR079583.

Author information

Author notes

  1. Matthew F. Mart and Joshua K. Johnson: Co-senior authors.

Authors and Affiliations

  1. Department of Physical Therapy, College of Health Sciences, University of Kentucky, 900 Rose Street, Lexington, KY, 40536, USA

    Kirby P. Mayer & Lindsey E. Fresenko

  2. University of Michigan Hospital, University of Michigan Health, Ann Arbor, MI, USA

    Evan Haezebrouck

  3. Division of Biokinesiology and Physical Therapy, University of Southern California, Los Angeles, CA, USA

    Lori M. Ginoza, Clarisa Martinez, Minnie Jan & Lori A. Michener

  4. Department of Internal Medicine, College of Medicine, University of Kentucky, Lexington, KY, USA

    Ashley A. Montgomery-Yates & Anna G. Kalema

  5. Division of Physical Therapy, Duke University School of Medicine, Durham, NC, USA

    Amy M. Pastva & Joshua K. Johnson

  6. Department of Physical Medicine and Rehabilitation, Cleveland Clinic, Cleveland, OH, USA

    Michelle Biehl & Joshua K. Johnson

  7. Division of Allergy, Pulmonary, and Critical Care, Vanderbilt University Medical Center, Nashville, TN, USA

    Matthew F. Mart

  8. Critical Illness, Brain Dysfunction, and Survivorship (CIBS) Center, Nashville, TN, USA

    Matthew F. Mart

  9. Geriatric Research, Education, and Clinical Center (GRECC), Tennessee Valley Veterans Affairs Healthcare System, Nashville, TN, USA

    Matthew F. Mart

Authors
  1. Kirby P. Mayer
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Evan Haezebrouck
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Lori M. Ginoza
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Clarisa Martinez
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Minnie Jan
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Lori A. Michener
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Lindsey E. Fresenko
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Ashley A. Montgomery-Yates
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Anna G. Kalema
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Amy M. Pastva
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Michelle Biehl
    View author publications

    You can also search for this author in PubMed Google Scholar

  12. Matthew F. Mart
    View author publications

    You can also search for this author in PubMed Google Scholar

  13. Joshua K. Johnson
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

All authors participated in conception of study, methodology and selection of variable for the database with regular meetings occurring in 2021 and 2022. KPM, EH, MFM, and JKJ performed initial data integrity, management, and statistical analyses. KPM drafted the manuscript and developed figures. All authors reviewed, edited, and provided final approval for the manuscript.

Corresponding author

Correspondence to Kirby P. Mayer.

Ethics declarations

Ethics approval and consent to participate

Medical IRB with expedited review was approved at University of Kentucky (#47751, 9/27/2022). Informed consent was waived due to the study design and deidentifed nature of data. Data User and Transfer Agreements were approved for sharing of data between sites using REDCap platform with data integrity checks.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Mayer, K.P., Haezebrouck, E., Ginoza, L.M. et al. Early physical rehabilitation dosage in the intensive care unit associates with hospital outcomes after critical COVID-19. Crit Care 28, 248 (2024). https://doi.org/10.1186/s13054-024-05035-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13054-024-05035-6

Keywords

Critical Care

ISSN: 1364-8535