Comments
Abstract
Rationale: The ICU Mobility Scale (IMS) is a measure of mobility milestones in critically ill patients.
Objectives: This study aimed to determine the validity and responsiveness of the IMS from a prospective cohort study of adults admitted to the intensive care unit (ICU).
Methods: Construct and predictive validity were assessed by comparing IMS values at ICU discharge in 192 patients to other variables using Spearman rank correlation coefficient, Mann-Whitney U tests, and logistic regression. Responsiveness was assessed using change over time, effect size, floor and ceiling effects, and percentage of patients showing change.
Measurements and Main Results: The IMS at ICU discharge demonstrated a moderate correlation with muscle strength (r = 0.64, P < 0.001). There was a significant difference between the IMS at ICU discharge in patients with ICU-acquired weakness (median, 4.0; interquartile range, 3.0–5.0) compared with patients without (median, 8.0; interquartile range, 5.0–8.0; P < 0.001). Increasing IMS values at ICU discharge were associated with survival to 90 days (odds ratio [OR], 1.38; 95% confidence interval [CI], 1.14–1.66) and discharge home (OR, 1.16; 95% CI, 1.02–1.32) but not with return to work at 6 months (OR, 1.09; 95% CI, 0.92–1.28). The IMS was responsive with a significant change from study enrollment to ICU discharge (d = 0.8, P < 0.001), with IMS values increasing in 86% of survivors during ICU admission. No substantial floor (14% scored 0) or ceiling (4% scored 10) effects were present at ICU discharge.
Conclusions: Our findings support the validity and responsiveness of the IMS as a measure of mobility in the ICU.
Clinical trial registered with www.clinicaltrials.gov (NCT01674608).
Appropriate intermediate, functional measures of limitations in physical function are required for clinicians and researchers in the intensive care unit (ICU) (1). Such endpoints should be valid, reliable, and responsive to change (2, 3). Ideally, the same endpoints should be used across studies to permit comparison of results (4). However, systematic reviews of physical endpoints and outcome measures in critical care research have found great variability in the measures used (5, 6). A systematic review completed in 2012 found that the most commonly used outcomes in rehabilitation trials in intensive care were the ability to perform mobility milestones such as sitting, standing, and walking (5). However, these were recorded in different ways, making it difficult to compare results (5). As a result, the ICU Mobility Scale (IMS) was developed to provide a structured method to collect mobility data, to assist clinicians monitoring recovery, to help researchers objectively quantify mobility milestones, and to compare the levels of mobility achieved in different studies (7).
The IMS provides a quick and simple bedside method of measuring a critically ill patient’s mobility (7). In contrast to mobility milestones (i.e., first time to stand or walk), which are commonly used as intermediate, functional endpoints in studies of rehabilitation in the ICU (5), the IMS provides a sensitive 11-point ordinal scale, ranging from nothing (lying/passive exercises in bed, score of 0) to independent ambulation (score of 10) (Table 1). The IMS has been used in observational studies (8, 9) and is currently being used in three studies registered in Australia and New Zealand.
| Classification | Definition | |
|---|---|---|
| 0 | Nothing (lying in bed) | Passively rolled or passively exercised by staff, but not actively moving |
| 1 | Sitting in bed, exercises in bed | Any activity in bed, including rolling, bridging, active exercises, cycle ergometry, and active assisted exercises; not moving out of bed or over the edge of the bed |
| 2 | Passively moved to chair (no standing) | Hoist, passive lift, or slide transfer to the chair, with no standing or sitting on the edge of the bed |
| 3 | Sitting over edge of bed | May be assisted by staff, but involves actively sitting over the side of the bed with some trunk control |
| 4 | Standing | Weight bearing through the feet in the standing position, with or without assistance. This may include use of a standing lifter device or tilt table. |
| 5 | Transferring bed to chair | Able to step or shuffle through standing to the chair. This involves actively transferring weight from one leg to another to move to the chair. If the patient has been stood with the assistance of a medical device, they must step to the chair (not included if the patient is wheeled in a standing lifter device). |
| 6 | Marching on spot (at bedside) | Able to walk on the spot by lifting alternate feet (must be able to step at least 4 times, twice on each foot), with or without assistance |
| 7 | Walking with assistance of 2 or more people | Walking away from the bed/chair by at least 5 m (5 yd) assisted by 2 or more people |
| 8 | Walking with assistance of 1 person | Walking away from the bed/chair by at least 5 m (5 yd) assisted by 1 person |
| 9 | Walking independently with a gait aid | Walking away from the bed/chair by at least 5 m (5 yd) with a gait aid, but no assistance from another person. In a wheelchair-bound person, this activity level includes wheeling the chair independently 5 m (5 yr) away from the bed/chair |
| 10 | Walking independently without a gait aid | Walking away from the bed/chair by at least 5 m (5 yd) without a gait aid or assistance from another person |
The IMS is feasible and reliable, with evidence of face and content validity (7). A recent study assessing the validity and responsiveness of various measures demonstrated that the IMS had criterion validity, was predictive of discharge destination, and showed change over time from awakening to ICU discharge (10).
As the IMS use in ICU research is increasing, it is important to further evaluate the utility of the IMS as a research and clinical tool in ICU patients. Therefore, our aim was to assess its generalizability, construct and predictive validity, and responsiveness (Table 2) using data collected in a large multicenter observational study. We hypothesized that the IMS at ICU discharge would be at least moderately correlated (r > 0.5) (11) with Medical Research Council Sum Score (MRC-SS) at ICU discharge and self-reported health (EQ-5D, analog scale) at 6 months and would not be correlated (r < 0.25) (11) with patient body weight.
| Type | Definition | Methods Used |
|---|---|---|
| Validity | The extent to which a tool measures what it is intended (3, 33) | IMS at ICU DC compared with the following variables: |
| Construct | ||
| Convergent | Measure of constructs that theoretically should be related to each other | Muscle strength (MRC-SS) |
| Divergent | Measure of constructs that theoretically should not be related to each other | Weight |
| Sex | ||
| Predictive | The ability for an outcome measure to predict a future outcome | 90-d mortality |
| Discharge destination | ||
| Return to work (6 mo) | ||
| EQ-5D analog scale (6 mo) | ||
| Responsiveness | The ability of an outcome measure to detect change (3) | |
| Change over time | Assesses the change in group results of a measure over various time points measured | Median difference in IMS values from enrollment to ICU DC and from ICD DC to 6 mo and effect sizes |
| Ability to detect change | Assesses the proportion of patients not showing change over various time points | Percentage of patients showing no change between enrollment to ICU DC and from ICD DC to 6 mo |
| Floor and ceiling effects | Determines the proportion of patients scoring the lowest and highest scores on the measure | Percentage of patients scoring 0 and 10 at enrollment and ICU DC |
Data were collected prospectively during an observational study conducted in 12 ICUs across Australia and New Zealand (8). Each participating center’s Human Research Ethics Committee approved this study, with either next of kin consent or a waiver of consent.
Patients were eligible for inclusion if they had independent mobility before hospital admission (as reported by next of kin) and were mechanically ventilated within the previous 24 hours, with the expectation of remaining mechanically ventilated in the ICU for another 48 hours. Patients were excluded if they were less than 18 years of age, were admitted with new onset of proven or suspected neurological impairment (i.e., stroke, traumatic brain injury, or Guillain-Barré syndrome), were unable to communicate in English, had cognitive impairment before the ICU admission (as reported by next of kin), had unstable fractures or injuries that would require bed rest orders, were receiving comfort measures only, or had a primary myopathic or neurological process associated with muscle weakness (e.g., Guillain-Barré syndrome).
The IMS was collected at enrollment to the study, at ICU discharge (by the treating physiotherapist), and at 6 months after enrollment (via telephone interview). Additional outcome measures evaluated in the study include: muscle strength at ICU discharge (assessed using the MRC-SS) (11), sedation during ICU (assessed using the Richmond Agitation Sedation Scale [RASS]) (12), hospital discharge destination (determined from the medical records and phone call to participant or next of kin), and 90-day mortality. Return to work and self-reported health via EQ-5D analog scale were collected at 6 months from enrollment in the study.
Construct and predictive validity of the IMS at ICU discharge were evaluated (Table 2). Construct validity was assessed by hypothesizing that the IMS would be more closely related to muscle strength measures and less related to weight and sex. This was assessed in two areas:
| 1. | Convergent validity: the correlation between the IMS and muscle strength MRC-SS at ICU discharge was assessed. | ||||
| 2. | Divergent validity: the IMS at ICU discharge was compared with variables that were hypothesized to be less closely related with mobility. Correlation between the IMS and weight was assessed and the IMS was compared between men and women. | ||||
| Type of Validity | Test | Hypothesis and Direction | Variable | N | Result r and 95% CI, IMS median (IQR), or OR, 95% CI | P Value | Hypothesis Met |
|---|---|---|---|---|---|---|---|
| Construct | |||||||
| Convergent | Spearman rank | Positive, moderate | MRC-SS | 87 | r = 0.64; 95% CI, 0.49 to 075 | <0.001* | Yes |
| Convergent | Mann-Whitney U test | Significant | ICU AW: Yes | 43 | 4.0 (3.0–5.0) | <0.001* | Yes |
| ICU AW: No | 44 | 8.0 (5.0–8.0) | |||||
| Divergent | Spearman rank | No correlation | Weight | 150 | r = 0.06, 95% CI, −0.10 to 0.22 | 0.47 | Yes |
| Divergent | Mann-Whitney U test | No correlation | Sex: Male | 93 | 5.0 (3.0–7.0) | 0.12 | Yes |
| Sex: Female | 61 | 4.0 (1.5–8.0) | |||||
| Predictive | Logistic regression† | Significant, positive | Alive at 90 d | 154 | OR, 1.38; 95% CI, 1.14 to 1.66 | 0.001* | Yes |
| Logistic regression† | Significant, positive | Discharged home | 133 | OR, 1.16; 95% CI, 1.02 to 1.32 | 0.03* | Yes | |
| Logistic regression† | Significant, positive | Return to work | 73 | OR, 1.09; 95% CI, 0.92 to 1.28 | 0.33 | No | |
| Spearman rank | Positive, moderate | EQ-5D analog scale | 110 | r = −0.03; 95% CI, −0.21 to 0.16 | 0.77 | No |
Responsiveness was evaluated by assessing change over time (14), effect sizes (15), and percentage of patients showing no change (Table 2). The IMS values were compared to assess significance of change over time from study enrollment to ICU discharge and ICU discharge to 6 months post enrollment. Effect sizes for the change in the IMS between the three time points were calculated. The percentage of patients showing no change was calculated between study enrollment and ICU discharge and ICU discharge and 6 months post enrollment.
Floor and ceiling effects were calculated by assessing the number and percentage of participants scoring the lowest value (0) and the highest value (10) at study enrollment and at ICU discharge, as the IMS was developed specifically for use in ICU. Floor and ceiling effects less than 15% were considered acceptable (2).
Statistical analysis was completed using SPSS Windows Version 22 (Chicago, IL). Data were assessed for normality using the Kolmogorov-Smirnov statistic. Descriptive statistics were reported as mean and SD for normally distributed data and median and interquartile range (IQR) for nonnormally distributed data. As the IMS has natural ordering, it is appropriate to use nonparametric methods, as these use ranking order rather than assuming that the intervals represented between each level of IMS are exactly the same in some quantitative sense.
Spearman rank correlation coefficient was used to assess correlation between the IMS values and continuous variables (MRC-SS, weight, and EQ-5D analog scale). Correlations were defined as excellent (≥0.75), moderate (0.50–0.74), fair (0.25–0.49), and no meaningful correlation (<0.25) (11). The difference between the IMS values in two categorical groups (i.e., sex) was assessed using Mann-Whitney U test. Predictive validity was assessed using logistic regression, adjusting for age, APACHE II, and FCI, presented as odds ratio (OR) and 95% confidence interval (CI). Change over time from study enrollment to ICU discharge and ICU discharge to 6 months was assessed using Wilcoxon signed rank test. Median and IQR for the IMS differences between these time points were calculated. Effect sizes (d) were calculated between the three time points; this was defined as r = Z divided by the square root of sample size (15). Interpretation of the change was defined as small (d = 0.2–0.49), moderate (d = 0.5–0.79), and large (d ≥ 0.8) (13). The proportion of patients showing no change was assessed for significance using the chi-square test. A two-sided P value ≤ 0.05 was considered to be statistically significant.
A total of 192 participants were enrolled in the study (Figure 1), with demographic data detailed in Table 4. Among the study population, the IMS values available for analysis at each time point were as follows: 188 at enrollment (98% of alive participants), 154 at ICU discharge (100% of alive participants), and 120 at 6-month follow up (87% of participants).
Figure 1. Flow of patients through the study. Reprinted by permission from Reference 8.
[More] [Minimize]| Baseline Data | Total (n = 192) |
|---|---|
| Age, mean (SD), yr | 58.0 (15.8) |
| Men, n (%) | 117 (61) |
| Previously walking independently, n (%) | 192 (100) |
| Previously working, n (%) | 77 (40.1) |
| Weight, mean (SD), kg | 85.1 (25.2) |
| APACHE II score, mean (SD) | 19.1 (7.6) |
| Functional Comorbidity Index (34), median (IQR) | 1 (1–2) |
| Vasoactive drugs, n (%) | 127 (68) |
| Sedation (RASS) n (%) | 124 (64) |
| Time ICU admission to enrollment, median (IQR), d | 2 (1–2) |
| Principal diagnoses, n (%) | |
| Cardiac or cardiothoracic | 59 (30) |
| Respiratory | 40 (20) |
| Gastrointestinal | 34 (17) |
| Sepsis | 28 (15) |
| Other | 31 (17) |
| ICU length of stay, median (IQR), d | 11 (6–17) |
| Hospital length of stay, median (IQR), d | 24 (16–42) |
Construct validity was demonstrated through a moderate correlation between the IMS and MRC-SS at ICU discharge (r = 0.64, P < 0.001) (Table 3) and significantly higher IMS values for participants without ICU-acquired weakness (16) (median, 8.0; IQR, 5.0–8.0) compared with those with ICU-acquired weakness (median, 4.0; IQR, 3.0–5.0; P < 0.001) (Table 3). No meaningful correlation was observed between the IMS at ICU discharge and the unrelated variable, weight (r = 0.06) (Table 3); and no significant difference was observed in the IMS at ICU discharge in men (median, 5.0; IQR, 3.0–7.0) compared with women (median, 4.0; IQR, 1.5–8.0; P = 0.12) (Table 3). There was also no significant difference in IMS values according to sex at enrollment (P = 0.34) or 6 months (P = 0.19).
Predictive validity was evident with increasing IMS values being associated with survival to 90 days (OR, 1.38; 95% CI, 1.14–1.66) and discharge home (OR, 1.16; 95% CI, 1.02–1.32), after adjusting for age, APACHE II, and FCI (Table 3). Table 5 outlines the proportion of patients discharged home based on their IMS values at ICU discharge. This shows that a greater proportion of patients with higher IMS values at ICU discharge are discharged directly home. The IMS at ICU discharge was not significantly associated with return to work at 6 months (OR, 1.09; 95% CI, 0.92–1.28), and there was no meaningful correlation between the IMS at ICU discharge and EQ-5D analog scale at 6 months (r = −0.03, P = 0.77) (Table 3).
| IMS at ICU Discharge | Discharged Home, n (%) |
|---|---|
| 0 | 5 (42) |
| 1 | 5 (83) |
| 2 | 6 (50) |
| 3 | 3 (30) |
| 4 | 7 (47) |
| 5 | 11 (50) |
| 6 | 6 (67) |
| 7 | 10 (71) |
| 8 | 16 (67) |
| 9 | 2 (50) |
| 10 | 5 (100) |
The effect sizes for change in the IMS from study enrollment (median, 0; IQR, 0–0) to ICU discharge (median, 5; IQR, 2–7) and from ICU discharge to 6 months (median, 10; IQR, 10–10) were large (d = 0.8 between each time point). Approximately 86% of survivors showed improvement in the IMS values between study enrollment and ICU discharge (P < 0.001), and 92% showed improvement between ICU discharge and 6-month follow up (P < 0.001).
The IMS showed acceptable floor (14% scored 0) and ceiling (4% scored 10) effects at ICU discharge; however, significant floor effects were present at study enrollment (96% scored 0), when patients were deeply sedated (median RASS score, −4).
Using multicenter observational data, this study has shown that the IMS at ICU discharge has evidence of construct validity, with measurement of muscle strength and ICU-acquired weakness and was predictive of 90-day mortality and discharge destination. The IMS was also responsive to change, with large effect sizes and no floor or ceiling effects present at ICU discharge. This is the first study to assess the predictive validity of the IMS in relation to 90-day mortality.
Other measures of physical function that have been designed for use in the ICU include (1) the Functional Status Score for the ICU, which is quick to complete and has promising validity and responsiveness data but has no reliability data in the ICU population (6, 17–19); (2) Physical Function in ICU Test (PFIT), which has good reliability and validity (with 6-min-walk test, MRC-SS, grip strength, and timed up and go test) but requires a stopwatch and can take up to 20 minutes to complete (6, 20, 21); and (3) the Chelsea Critical Care Physical Assessment tool (CPAx), with established reliability, validity (with Short Form-12 Physical Component Score), correlation with hospital discharge destination, and responsiveness (22). The IMS has the benefit of being simple to score, being reliable when completed by either nurses or physiotherapists, taking less than 1 minute to complete, and requiring no specific equipment (7). None of the other measures of mobility or function have been tested as to whether they are predictive of 90-day mortality, whereas this has been shown with the IMS in this study.
The MRC-SS is a measure of muscle strength, which has been used in many ICU trials and is used in the definition of ICU-acquired weakness (19, 23–25). The MRC-SS has been shown to have a significant association with ICU and hospital length of stay (26). There is conflicting evidence on the association between muscle strength and mortality (26, 27). The 2-minute-walk test and the 6-minute-walk test are commonly used measures of exercise tolerance in many patient populations (28, 29). These tests have been used in ICU rehabilitation studies (24, 30, 31); however, they are mostly collected at hospital discharge or well after ICU discharge and may not be feasible for use while patients are in the ICU (24, 32).
The construct and predictive validity properties of the IMS support the use of the IMS in the ICU to measure patients’ daily mobility level. In this study, the IMS at ICU discharge was predictive of 90-day mortality and discharge destination. The significance of poor mobility on patient survival supports the need for large studies to assess the use of early goal-directed mobility to improve patient outcome.
This study was conducted in 12 ICUs in Australia and New Zealand and includes both large and small ICUs in urban centers, providing a degree of external validity to our study findings. Our sample size of 192 is large in comparison to other validation studies completed in ICU patients (6, 17, 21). Other strengths of this study include low rate of loss to follow up, central assessment of long-term outcomes, and broad eligibility criteria. Detailed demographic data were reported, allowing comparison of the study population to other populations.
This study also carries some limitations. This study only included patients who were independently mobile before ICU admission, and there was a long ICU and hospital length of stay. Therefore, it is unsure whether these results could be generalized to a population of patients who had poor mobility before ICU admission or have short ICU and hospital length of stay. There may be some concerns regarding the substantial floor effects measured at enrollment to the study. However, the IMS was developed by both clinicians and researchers to ensure that each category on the scale was relevant and was associated with a key mobility milestone (7). The floor effects present in this analysis are likely related to the deep sedation (and corresponding low RASS levels) of participants on enrollment to the study—accurately representing uniform deep sedation on study Day 1 rather than representing inability to detect meaningful variation in activity that was somehow present.
As this study used data from an observational study, it was not able to specifically assess all aspects of validity and responsiveness. Future studies are warranted to further assess the validity, particularly in relation to 90-day mortality and discharge destination and to determine a minimal important difference. Eighteen patients were lost to follow up. No assumptions were made for these patients, and therefore no imputations were made; this may affect predictive validity in relation to 90-day mortality. As height was not collected, divergent validity was only able to be assessed regarding weight; it would be useful to assess this regarding body mass index. The IMS has not been assessed for reliability via self-reporting; future studies are warranted to assess this at various time points of patient recovery.
In this study, the IMS was not predictive of return to work or self-reported health. The population studied had low prior working levels (40%) and therefore due to low numbers the study may not have had adequate power to determine predictive validity. This needs to be assessed in a larger cohort. It was hypothesized that there would be a moderate correlation between the IMS at ICU discharge and self-reported health at 6 months; however, this was not the case in this study population. To further assess the relationship between mobility and quality of life, it would be useful to collect a data set of the IMS and a variety of quality-of-life measures collected at similar time points.
The IMS demonstrated construct validity, was predictive of 90-day mortality and discharge destination, and was responsive in an ICU population. This simple scale of mobility adds to the list of available endpoints used to assess mobility and function in ICU patients. The IMS is useful in providing a standardized method for assessing the daily highest level of mobilization in the ICU, for clinical and research purposes.
| 1 . | Young P, Hodgson C, Dulhunty J, Saxena M, Bailey M, Bellomo R, Davies A, Finfer S, Kruger P, Lipman J, et al.; ANZICS Clinical Trials Group. End points for phase II trials in intensive care: recommendations from the Australian and New Zealand Clinical Trials Group consensus panel meeting. Crit Care Resusc 2012;14:211–215. |
| 2 . | Terwee CB, Bot SD, de Boer MR, van der Windt DA, Knol DL, Dekker J, Bouter LM, de Vet HC. Quality criteria were proposed for measurement properties of health status questionnaires. J Clin Epidemiol 2007;60:34–42. |
| 3 . | Roach K. Measurement of health outcomes: reliability. Validity and Responsiveness JPO 2006;18:8. |
| 4 . | Williamson P, Clarke M. The COMET (Core Outcome Measures in Effectiveness Trials) initiative: its role in improving Cochrane Reviews. Cochrane Database Syst Rev 2012;5:ED000041. |
| 5 . | Tipping CJ, Young PJ, Romero L, Saxena MK, Dulhunty J, Hodgson CL. A systematic review of measurements of physical function in critically ill adults. Crit Care Resusc 2012;14:302–311. |
| 6 . | Parry SM, Granger CL, Berney S, Jones J, Beach L, El-Ansary D, Koopman R, Denehy L. Assessment of impairment and activity limitations in the critically ill: a systematic review of measurement instruments and their clinimetric properties. Intensive Care Med 2015;41:744–762. |
| 7 . | Hodgson C, Needham D, Haines K, Bailey M, Ward A, Harrold M, Young P, Zanni J, Buhr H, Higgins A, et al. Feasibility and inter-rater reliability of the ICU Mobility Scale. Heart Lung 2014;43:19–24. |
| 8 . | Hodgson C, Bellomo R, Berney S, Bailey M, Buhr H, Denehy L, Harrold M, Higgins A, Presneill J, Saxena M, et al.; TEAM Study Investigators. Early mobilization and recovery in mechanically ventilated patients in the ICU: a bi-national, multi-centre, prospective cohort study. Crit Care 2015;19:81. |
| 9 . | Ma AJ, Zhan A, Sigurdsson G, Rawat N, Needham D, Saria S. Feasibility of a non-invasive sensore for measuring ICU patient mobility [abstract]. Crit Care Med 2014;42:A1389. |
| 10 . | Parry SM, Denehy L, Beach LJ, Berney S, Williamson HC, Granger CL. Functional outcomes in ICU: what should we be using? An observational study. Crit Care 2015;19:127. |
| 11 . | Fan E, Ciesla ND, Truong AD, Bhoopathi V, Zeger SL, Needham DM. Inter-rater reliability of manual muscle strength testing in ICU survivors and simulated patients. Intensive Care Med 2010;36:1038–1043. |
| 12 . | Shehabi Y, Bellomo R, Reade MC, Bailey M, Bass F, Howe B, McArthur C, Seppelt IM, Webb S, Weisbrodt L; Sedation Practice in Intensive Care Evaluation (SPICE) Study Investigators; ANZICS Clinical Trials Group. Early intensive care sedation predicts long-term mortality in ventilated critically ill patients. Am J Respir Crit Care Med 2012;186:724–731. |
| 13 . | Cohen J. Statistical power analysis for the behavioural sciences, 2nd ed., Hillsdale, NJ: Lawrence Erlbaum Associates; 1988. |
| 14 . | Deyo RA, Centor RM. Assessing the responsiveness of functional scales to clinical change: an analogy to diagnostic test performance. J Chronic Dis 1986;39:897–906. |
| 15 . | Portney L, Watkins M. Foundations of clinical research: applications to practise, 3rd ed. East Norwalk, CT: Appleton & Lange; 2009. |
| 16 . | De Jonghe B, Sharshar T, Lefaucheur JP, Authier FJ, Durand-Zaleski I, Boussarsar M, Cerf C, Renaud E, Mesrati F, Carlet J, et al.; Groupe de Réflexion et d’Etude des Neuromyopathies en Réanimation. Paresis acquired in the intensive care unit: a prospective multicenter study. JAMA 2002;288:2859–2867. |
| 17 . | Thrush A, Rozek M, Dekerlegand JL. The clinical utility of the functional status score for the intensive care unit (FSS-ICU) at a long-term acute care hospital: a prospective cohort study. Phys Ther 2012;92:1536–1545. |
| 18 . | Zanni JM, Korupolu R, Fan E, Pradhan P, Janjua K, Palmer JB, Brower RG, Needham DM. Rehabilitation therapy and outcomes in acute respiratory failure: an observational pilot project. J Crit Care 2010;25:254–262. |
| 19 . | Kho ME, Truong AD, Zanni JM, Ciesla ND, Brower RG, Palmer JB, Needham DM. Neuromuscular electrical stimulation in mechanically ventilated patients: a randomized, sham-controlled pilot trial with blinded outcome assessment. J Crit Care 2015;30:32–39. |
| 20 . | Skinner EH, Berney S, Warrillow S, Denehy L. Development of a physical function outcome measure (PFIT) and a pilot exercise training protocol for use in intensive care. Crit Care Resusc 2009;11:110–115. |
| 21 . | Denehy L, de Morton NA, Skinner EH, Edbrooke L, Haines K, Warrillow S, Berney S. A physical function test for use in the intensive care unit: validity, responsiveness, and predictive utility of the physical function ICU test (scored). Phys Ther 2013;93:1636–1645. |
| 22 . | Corner EJ, Soni N, Handy JM, Brett SJ. Construct validity of the Chelsea critical care physical assessment tool: an observational study of recovery from critical illness. Crit Care 2014;18:R55. |
| 23 . | Winkelman C, Johnson KD, Hejal R, Gordon NH, Rowbottom J, Daly J, Peereboom K, Levine AD. Examining the positive effects of exercise in intubated adults in ICU: a prospective repeated measures clinical study. Intensive Crit Care Nurs 2012;28:307–318. |
| 24 . | Nordon-Craft A, Schenkman M, Ridgeway K, Benson A, Moss M. Physical therapy management and patient outcomes following ICU-acquired weakness: a case series. J Neurol Phys Ther 2011;35:133–140. |
| 25 . | Schweickert WD, Pohlman MC, Pohlman AS, Nigos C, Pawlik AJ, Esbrook CL, Spears L, Miller M, Franczyk M, Deprizio D, et al. Early physical and occupational therapy in mechanically ventilated, critically ill patients: a randomised controlled trial. Lancet 2009;373:1874–1882. |
| 26 . | Connolly BA, Jones GD, Curtis AA, Murphy PB, Douiri A, Hopkinson NS, Polkey MI, Moxham J, Hart N. Clinical predictive value of manual muscle strength testing during critical illness: an observational cohort study. Crit Care 2013;17:R229. |
| 27 . | Ali NA, O’Brien JM Jr, Hoffmann SP, Phillips G, Garland A, Finley JC, Almoosa K, Hejal R, Wolf KM, Lemeshow S, et al.; Midwest Critical Care Consortium. Acquired weakness, handgrip strength, and mortality in critically ill patients. Am J Respir Crit Care Med 2008;178:261–268. |
| 28 . | ATS Committee on Proficiency Standards for Clinical Pulmonary Function Laboratories. ATS statement: guidelines for the six-minute walk test. Am J Respir Crit Care Med 2002;166:111–117. |
| 29 . | Holland AE, Hill CJ, Rasekaba T, Lee A, Naughton MT, McDonald CF. Updating the minimal important difference for six-minute walk distance in patients with chronic obstructive pulmonary disease. Arch Phys Med Rehabil 2010;91:221–225. |
| 30 . | Burtin C, Clerckx B, Robbeets C, Ferdinande P, Langer D, Troosters T, Hermans G, Decramer M, Gosselink R. Early exercise in critically ill patients enhances short-term functional recovery. Crit Care Med 2009;37:2499–2505. |
| 31 . | Denehy L, Skinner EH, Edbrooke L, Haines K, Warrillow S, Hawthorne G, Gough K, Hoorn SV, Morris ME, Berney S. Exercise rehabilitation for patients with critical illness: a randomized controlled trial with 12 months of follow-up. Crit Care 2013;17:R156. |
| 32 . | Chan KS, Pfoh ER, Denehy L, Elliott D, Holland AE, Dinglas VD, Needham DM. Construct validity and minimal important difference of 6-minute walk distance in survivors of acute respiratory failure. Chest 2015;147:1316–1326. |
| 33 . | Mokkink LB, Terwee CB, Patrick DL, Alonso J, Stratford PW, Knol DL, Bouter LM, de Vet HCW. The COSMIN checklist for assessing the methodological quality of studies on measurement properties of health status measurement instruments: an international Delphi study. Qual Life Res 2010;19:539–549. |
| 34 . | Fan E, Gifford JM, Chandolu S, Colantuoni E, Pronovost PJ, Needham DM. The functional comorbidity index had high inter-rater reliability in patients with acute lung injury. BMC Anesthesiol 2012;12:21. |
Supported by an Alfred Hospital Physiotherapy Research Fellowship and The Intensive Care Foundation of Australia and New Zealand.
Author Contributions: All authors contributed to conception and design, acquisition of data, and drafting the article or revising it critically for important intellectual content; C.J.T., C.L.H., and M.J.B. contributed to analysis and interpretation of data; and all authors had final approval of the version to be published.
Author disclosures are available with the text of this article at www.atsjournals.org.
