Physiotherapy 99 (2013) 33–41

The Chelsea Critical Care Physical Assessment Tool (CPAx): validation of

an innovative new tool to measure physical morbidity in the general adult
critical care population; an observational proof-of-concept pilot study
E.J. Corner a,∗ , H. Wood a , C. Englebretsen a , A. Thomas b , R.L. Grant c,d ,
D. Nikoletou c,d , N. Soni a
a Chelsea and Westminster NHS Foundation Trust, 369 Fulham Road, London SW10 9NH, UK
b The Royal London Hospital, Barts and The London NHS Trust, London, UK
c Faculty of Health and Social Care Sciences, Kingston University, Kingston, UK
d St. George’s Hospital Medical School, University of London, London, UK

Abstract
Objective To develop a scoring system to measure physical morbidity in critical care – the Chelsea Critical Care Physical Assessment Tool
(CPAx).
Method The development process was iterative involving content validity indices (CVI), a focus group and an observational study of 33
patients to test construct validity against the Medical Research Council score for muscle strength, peak cough flow, Australian Therapy
Outcome Measures score, Glasgow Coma Scale score, Bloomsbury sedation score, Sequential Organ Failure Assessment score, Short Form
36 (SF-36) score, days of mechanical ventilation and inter-rater reliability.
Participants Trauma and general critical care patients from two London teaching hospitals.
Results Users of the CPAx felt that it possessed content validity, giving a final CVI of 1.00 (P < 0.05). Construct validation data showed
moderate to strong significant correlations between the CPAx score and all secondary measures, apart from the mental component of the
SF-36 which demonstrated weak correlation with the CPAx score (r = 0.024, P = 0.720). Reliability testing showed internal consistency of
α = 0.798 and inter-rater reliability of κ = 0.988 (95% confidence interval 0.791 to 1.000) between five raters.
Conclusion This pilot work supports proof of concept of the CPAx as a measure of physical morbidity in the critical care population, and is
a cogent argument for further investigation of the scoring system.
© 2012 Chartered Society of Physiotherapy. Published by Elsevier Ltd. All rights reserved.

Keywords: Rehabilitation; Critical illness; Physiotherapy; Muscular diseases; Outcome assessment

Introduction recommends early identification of patients at risk of physical

morbidity by regular short clinical assessments at intervals
Critical illness frequently results in impaired neuromuscu- throughout the patient’s hospital stay. Once identified as
lar function and hence debilitation [1–4]. This phenomenon ‘at risk’, a comprehensive physical assessment should be
is commonly known as intensive care unit acquired weakness completed, and a structured patient-specific rehabilitation
(ICU-AW). programme commenced. Progress should be monitored using
In March 2009, the UK National Institute for Health patient-agreed goals and should be reviewed regularly.
and Clinical Excellence (NICE) introduced Clinical Guide- NICE CG83 was a long-awaited guideline and a mile-
line 83 (CG83), entitled ‘Rehabilitation after critical illness’, stone in the management of ICU-AW. However, the document
which aims to optimise the management of ICU-AW [5]. It highlights the lack of validated assessment tools and paucity
of evidence-based rehabilitation practices in the critically ill
∗ population. For example, Section 2.1 (p. 23) of CG83 focuses
Corresponding author. Tel.: +0208 7465689; fax: +0208 8466069.
E-mail address: [email protected] (E.J. Corner). on identifying screening and assessment tools to assess

0031-9406/$ – see front matter © 2012 Chartered Society of Physiotherapy. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.physio.2012.01.003
34 E.J. Corner et al. / Physiotherapy 99 (2013) 33–41

physical morbidity. Only one research article [6] was consid- The CPAx was developed using a pragmatic iterative
ered to be sufficiently scientifically robust for inclusion; this process recommended by Streiner and Norman [12] and
looked at the Rivermead Motor Index and even this ‘should be mimicking a modified Delphi technique (Fig. 1, see online
interpreted with caution due to its ill-defined inclusion crite- supplementary material). The final version of the CPAx is
ria and small population’. The lack of a validated assessment shown in Appendix 1.
tool poses a significant obstacle to the practical application Supplementary material related to this article found, in the
of CG83, preventing its implementation in a standardised online version, at doi:10.1016/j.physio.2012.01.003.
manner at a national level. Further implications include the
lack of an objective marker for standardisation of patient
management and health economic assessment. It also has Stage 1. Initial development of the CPAx
implications in the research setting.
The most commonly accepted scoring system for the The CPAx was initially designed as a pictorial composite
assessment of muscle function is the Medical Research Coun- of 10 commonly assessed components of physical ability,
cil (MRC) score for muscle strength [7]. This score looks each graded on a six-point Guttman scale from complete
solely at strength, and as patients may learn to be functional dependence to independence. The initial working draft was
in the absence of strength, its generalisability and valid- reviewed internally by clinicians (medics, nurses and ther-
ity for measuring physical morbidity is questionable. More apists), and trialled informally within the critical care unit
importantly, it is time consuming and relies on a degree of (CCU). The feedback was used to modify the CPAx draft.
co-operation and cognitive function. From the patient per- This was repeated seven times to produce a final working
spective, it may be considered relatively abstract and difficult version.
to interpret. Other measures that have been trialled, such as
the Barthel Index and the Functional Independence Measure,
lack sensitivity in patients with low levels of function, and/or Stage 2. Focus group
they have not been validated or reliability tested in a critically
ill population [8,9]. The next stage was a focus group with a purposive sam-
The most realistic attempts to address the lack of validated ple of seven [including six clinical specialist/lead (Band 8a)]
assessment tools in critical care are the Physical Functional physiotherapists in critical care and acute rehabilitation from
ICU Test (PFIT) [10] and the University of Rochester Acute four London hospitals and the primary developer. A Band 8a
Care Evaluation (URACE) [11]. The PFIT relies on the physiotherapist has highly developed specialist knowledge
patient being able to stand up from a chair and march on and skills, and leads in a specialist clinical area, clinical audit
the spot. These tasks may be difficult for patients in the acute and research [13]. In total, the participants had 87 years of
phases of critical illness, and published results only exist for postgraduate experience, one postgraduate certificate, three
a very small sample size (n = 13). The URACE is a scoring master’s degrees, and one PhD in progress.
system that grades people’s independence with bed mobil- The aim of the focus group was to further develop the
ity, transfers (bed to chair), locomotion and stairs. It has not CPAx and assess its face and content validity. The focus group
been tested for reliability and validity, and no patient data was recorded and transcribed verbatim, and the principles of
have been published to date. It was also developed with input thematic content analysis were applied to justify subsequent
from clinicians at a single hospital site in the USA; thus, its changes.
generalisability is questionable. Table 1 summarises the focus group transcript, listing and
There is an obvious and urgent need for a practical, justifying the components of the CPAx.
holistic and reproducible bedside measure of physical mor- Prior to the focus group, attendees reviewed the CPAx and
bidity. This could facilitate patient assessment, highlighting completed a Likert scale questionnaire to establish a content
problem areas and guiding rehabilitation; help to monitor validity index (CVI) [14,15]; a measure of the proportion of
progress; and form an objective measurement that is intelli- experts endorsing the CPAx (Appendix 2, see online supple-
gible between clinical specialties and service users. mentary material). Each question was scored out of four, and
The aim of this pilot study was to develop a bedside scor- a score of three or more for each component indicated that
ing system to grade physical morbidity in the critical care the expert endorsed that component. The CVI was equated
population – the Chelsea Critical Care Physical Assessment by dividing the number of experts endorsing the tool by the
Tool (CPAx). total number of experts. It was piloted on one physiotherapist,
whose data were not included in the study results. Following
the focus group, the CVI questionnaire was repeated with
Methods reference to the modified version of the CPAx, which was
used as a method of respondent validation and considered
This study involved several stages. For clarity, this paper the outcome of the focus group.
is presented with the method followed by the results for each Supplementary material related to this article found, in the
stage in order of completion. online version, at doi:10.1016/j.physio.2012.01.003.
 E.J. Corner et al. / Physiotherapy 99 (2013) 33–41 35

Table 1
Focus group summary including components of the Chelsea Critical Care Physical Assessment Tool (CPAx) and rationale for their inclusion.
Component of physicality Description Rationale for inclusion
Respiratory function Amount of respiratory support required, in terms of ICU-AW weakness is commonly associated with
both ventilation and oxygenation prolonged weaning from mechanical ventilation and
impaired cough function due to respiratory muscle
atrophy [25]. Inclusion of the cough and respiratory
function sections therefore gives a much more
holistic impression of the patient’s physical
problems associated with ICU-AW
Cough Cough effectiveness, in terms of consistency and
secretion clearance
Bed mobility The ability and level of assistance required to move All physical tasks with composite parts of
around the bed measurement scales and knowledge of the
functional impairment commonly associated with
ICU-AW. As the purpose of the CPAx is not only to
measure physical ability but also to act as an
outcome measure for physiotherapy research, it was
important to include components that would be
influenced by physiotherapy treatment
Supine to sitting on the edge of the bed The ability and level of assistance required to sit on
the edge of the bed from supine
Dynamic sitting The level of support required to maintain sitting
balance, progressing to the ability to reach out of the
base of support
Sit to stand The ability and level of assistance needed to stand
from a sitting position of >90◦ of knee flexion
Standing balance The amount of support required to maintain
standing, ranging from a tilt table to independent
Transferring from bed to chair Mode of transfer from bed to chair (e.g. cradle hoist
transfers, standing hoist, independent, etc.)
Stepping Assesses the physical ability to walk, and support
required (e.g. frame, physical assistance, etc.)
Grip strength Grip strength measured by a grip dynamometer as a Grip strength has been shown to accurately reflect
percentage of expected, when age and gender whole-body strength and has also been used as a
corrected diagnostic tool for ICU-AW [19]
ICU-AW, intensive care unit acquired weakness.

Stage 2. Results Stage 4. Observational cohort study

The pre focus group CVI was 0.67 (P > 0.05) and the Study design
post focus group CVI was 0.83 (P < 0.05), indicating that This was a prospective cohort study of general and trauma
the CPAx was considered, by a small number of specialist CCU patients admitted to Chelsea and Westminster NHS
clinicians, to be content valid. Foundation Trust and the Royal London Hospital NHS Trust
between June and August 2010. Following recruitment, all
Stage 3. CVI questionnaire participants were assessed by a physiotherapist who used the
CPAx on admission, discharge and every Monday, Wednes-
To further explore the content validity of the CPAx, it day and Friday for the duration of their CCU stay. One
was distributed for use across three UK teaching hospitals. retrospective CPAx score was collected as a measure of
Following a minimum 3-month trial, the CVI questionnaire premorbid physical level (this excluded the grip strength
was sent out to all physiotherapists who had been using the component). The scores were blinded to the researcher who
tool regularly (n = 14); clinicians included in the development collected further data (listed below) to test the construct valid-
of the tool were excluded. ity of the CPAx. These measures were taken on the same day
as each CPAx score:
Stage 3. Results
• Peak cough flow;
Fourteen clinicians responded to the questionnaire and all • Australian Therapy Outcome Measures (AusTOMs) score;
of them rated the tool as content valid, giving a final CVI of • Medical Research Council (MRC) score for quadriceps
1.00 (P < 0.05). and biceps strength;
36 E.J. Corner et al. / Physiotherapy 99 (2013) 33–41

• Sequential Organ Failure Assessment (SOFA) score; muscle function from no muscle activity to full strength. The
• Bloomsbury sedation scale (scores −3 to 1); MRC score was selected as it is one of the most common
• Glasgow Coma Scale (GCS) score; objective measurement scales used by physiotherapists on
• Number of days of mechanical ventilation; and the CCU, as well as a diagnostic tool for ICU-AW [19].
• Short Form 36 (SF-36) score. Biceps and quadriceps muscle strength were chosen as they
are key anti-gravity muscles necessary for functional tasks,
As all of the above measures were taken by the researcher,
and completing a whole-body strength test would have been
no inter-rater reliability testing was required.
impractical [19,20]. For analysis, these scores were compared
in isolation and as a total out of 20.
Construct validation The measurements were taken in a seated position (either
A series of null hypotheses were constructed to ascertain in the electric bed or in the chair). Participants were asked
correlations between the CPAx and other variables measuring to resist the researcher from moving the limb. The limb was
the ‘physical morbidity’ construct. then graded on the scale, and data were recorded. If the par-
ticipant was unable to follow commands, no score was given;
Peak cough flow. Peak cough flow was compared with the however, if they were sedated, they scored 0.
cough component of the CPAx, as it is an objective measure
of cough strength [16]. Patients that could be disconnected SOFA score. The SOFA score was chosen as a compara-
from the ventilator (as determined by the clinical team) had tor as greater acuity of illness means a greater likelihood
their peak cough flow measured using either a calibrated Vita- of prolonged bed rest, and possibly more sedation and neu-
lograph Spirometer 2120 (Vitalograph, Buckingham, UK) or romuscular blocking agents, increasing the risk of ICU-AW
a calibrated Microplus Spirometer (Micro Medical Ltd, Car- [2,4].
dinal Health, Basingstoke, UK). Due to infection policy and
financial reasons, it was not possible to standardise equipment GCS and Bloomsbury sedation scale scores. The CPAx was
between sites. developed to be responsive from complete dependence to
If the patient could not be disconnected from the venti- independence, and thus needed to be validated throughout its
lator, the expiratory flow wave form on the ventilator was entirety. Other scores of physical morbidity have a significant
used to measure peak cough flow. An Evita XL ventilator floor effect in the CCU population, and therefore would be
(Dräger, Telford, PA, USA) was used at the lead site, and an poor comparators to the CPAx in the low-functioning patient.
Engstrom Carestation ventilator (GE Healthcare, Hatfield, As a result, both GCS and sedation scores were selected as
UK) was used at the second site. comparators. It was anticipated that both the GCS and seda-
Peak cough flow was measured in a standardised seated tion scores would only be comparable with the CPAx at the
position (i.e. 90◦ of knee flexion and hip flexion). Patients lower end of the CPAx scale. Sedation scores of 1 to −3
were given one practice trial followed by one measurement were chosen to allow correlation analysis by ensuring a linear
per session. All measurements were taken during sponta- relationship between the two measures.
neous coughing. If it was clinically inappropriate to measure
peak cough flow, the measurement was not taken. It is Number of days of mechanical ventilation. A longer time
accepted that the validity of the peak cough flow measurement on mechanical ventilation is associated with poorer func-
may be affected by different assessment techniques. tional outcome from critical illness [21]. Therefore, length
The remaining measures were compared with the total of time on mechanical ventilation was compared with the
CPAx score. CCU discharge CPAx score.

AusTOMs score. The AusTOMs is a generic Australian ther- SF-36 score. The SF-36 is a validated health-related
apy outcome measure [17,18] that grades patients on a quality-of-life questionnaire [22]. It is separated into two
Guttman scale (0 to 5) from complete dependence to inde- components: physical and mental. SF-36 scores were col-
pendence. To standardise the measurements, the activity lected to compare with the pre-admission CPAx scores.
limitation and impairment sections of both the balance and Furthermore, as recovery from critical illness is associated
postural control component and the musculoskeletal compo- with pre-admission health status [23,24], the SF-36 score was
nent of the AusTOMs were chosen for assessment. also compared with the CCU discharge CPAx score.
The AusTOMs comparator was selected because it was All statistical analyses were performed using Statistical
being used at both sites already, and is a simple measure that Package for the Social Sciences Version 16.0 (SPSS, IBM
can be graded easily by the researcher. The physiotherapy Corporation, NY, USA). Spearman’s rank correlation coeffi-
section of the AusTOMs has also demonstrated both face cient was used to analyse the construct validity of the CPAx
and content validity [17]. against the MRC score, SOFA score, AusTOMs score, seda-
tion score, GCS score and peak cough flow, as data were
Quadriceps and biceps muscle strength (MRC score). The skewed. One randomly selected time point for each partic-
MRC score is a six-point scoring system (0 to 5) that grades ipant was used in the correlation analysis, giving a total of
 E.J. Corner et al. / Physiotherapy 99 (2013) 33–41 37

Table 3
Construct validity results summary: Spearman’s rank correlation coefficients between Chelsea Critical Care Physical Assessment Tool (CPAx) scores and each
comparator.
Comparator R P-value
Peak cough flow 0.633 0.006
AusTOMs MSK score (activity) 0.735 <0.001
AusTOMs MSK score impairment 0.763 <0.001
AusTOMs BPC score (activity) 0.903 <0.001
AusTOMs BPC score (impairment) 0.874 <0.001
MRC score (right biceps) 0.693 <0.001
MRC score (left biceps) 0.640 <0.001
MRC score (right quads) 0.697 <0.001
MRC score (left quads) 0.673 <0.001
MRC total score 0.650 <0.001
SOFA score −0.683 <0.001
GCS score 0.764 <0.001
Bloomsbury sedation scale score (−3 to 1) 0.420 0.036
Days of mechanical ventilation −0.506 <0.01
SF-36 (physical component) and pre-admission CPAx score 0.720 0.013
SF-36 (mental component) and pre-admission CPAx score 0.122 0.720
SF-36 (mental component) and discharge CPAx score 0.024 0.954
SF-36 (physical component) and discharge CPAx score 0.843 0.009
AusTOMs, Australian Therapy Outcome Measures; MSK, musculoskeletal; BPC, balance and postural control; MRC, Medical Research Council; SOFA,
Sequential Organ Failure Assessment; GCS, Glasgow Coma Scale; SF-36, Short Form 36.

33 for each measure. This was because each individual par- functional deficit were excluded. The remaining exclusion
ticipant’s scores could naturally correlate with themselves. criteria were: expected length of stay <48 h (to eliminate rou-
Randomly analysing one time point for each participant elim- tine postoperative admissions), unable to achieve consent,
inated this risk. pregnancy, and if it was deemed inappropriate by the clinical
team.
Inter-rater reliability
Five physiotherapists each assessed three patients selected
Stage 4. Results
at random. The patient’s lead physiotherapist carried out the
assessment, which was observed by four other physiothera-
Participant characteristics
pists. The lead physiotherapist did not verbalise any of their
Seventy-six patients were identified for recruitment
assessment to the observers. Each physiotherapist then scored
into the study. Forty-three prospective participants either
the patient in isolation. All assessors were blinded to the other
declined, died, transferred out of the unit, or their clinical
scores. Intraclass correlation coefficients between each com-
picture changed making them inappropriate for the study.
ponent of the CPAx, as well as the total score, were equated.
Thirty-three participants were recruited. Participant attri-
Diagnosis, medication, age, past medical history,
tion and the consent procedure is displayed in Fig. 2 (see
APACHE II scores and outcome data were collected for
online supplementary material). The baseline demograph-
demographic purposes.
ics for these participants are shown in Table 2 (see online
supplementary material).
Study population
Supplementary material related to this article found, in the
A sample size calculation was completed for a Pearson’s
online version, at doi:10.1016/j.physio.2012.01.003.
correlation with a two-sided test, alpha = 0.05 and a power of
80% (1 − β). At least 26 participants were required to find
a statistically significant correlation of above 0.5 between Construct validity
MRC score and CPAx score. Thirty-three participants were One hundred and ninety-two CPAx scores were collected,
recruited to account for attrition. with a mean of six scores per patient. Peak cough flow was
measured successfully on 58% of occasions (n = 82). Rea-
Exclusion criteria sons for not measuring peak cough flow were unable to
The purpose of the CPAx score was to fill the void in the cough, sedated, on high-frequency oscillatory ventilation or
assessment of CCU patients in conjunction with NICE CG83, refusal. MRC score was measured successfully on 80% of
which is aimed at patients who do not fall under specialist occasions (n = 129). Reasons for not measuring MRC score
rehabilitation services. Therefore, cardiothoracic and burns were inability to follow commands/drowsiness or refusal. All
patients, and patients admitted with a focal neurology causing other measures were recorded successfully.
38 E.J. Corner et al. / Physiotherapy 99 (2013) 33–41

Spearman’s rank correlation coefficient analysis demon- There has been no attempt to weight the components of the
strated moderate to strong significant positive correlations CPAx. Weighting would make the CPAx psychometrically
between the CPAx score and the MRC score, GCS score, stronger, but at the expense of simple, quick administra-
sedation score −3 to 1, peak cough flow and AusTOMs tion and hence clinical utility. A psychometrically strong
score. There was a significant positive correlation between assessment that is never used due to complex administra-
pre-admission CPAx score/CCU discharge CPAx score, and tion is futile. Furthermore, as the CPAx spans the continuum
the physical component of the SF-36, but no correlation with from dependence to independence, the importance of each
the mental component of the SF-36. There was a significant component will differ depending on the recovery stage.
negative correlation between CPAx score and SOFA score, What is clinically significant in the acute stages of illness
and CCU discharge CPAx score and the number of days will differ with rehabilitation stage. Therefore, weighting
of mechanical ventilation. A summary of these statistics is would be difficult, if not impossible. The consequence
presented in Table 3. is that the CPAx is appropriately regarded as an ordinal
scale.
Reliability
Inter-rater reliability and internal consistency (Cronbach’s Limitations
alpha) were strong at κ = 0.988 (95% confidence interval
0.791 to 1.000; P < 0.01) and α = 0.798, respectively. A sum- In this report, the sample size is limited, both in the focus
mary of the intraclass correlation coefficients for inter-rater group and the cohort study. There was an insignificant corre-
reliability is presented in Table 4 (see online supplementary lation between CPAx score and the mental component of the
material). SF-36. As the CPAx is designed to assess physical morbid-
Supplementary material related to this article found, in the ity, the non-significant correlation with mental health status
online version, at doi:10.1016/j.physio.2012.01.003. is unsurprising.
The CPAx was developed in London, so generalisabil-
ity may be limited. Further assessment in other centres is
Discussion required, but it is thought unlikely that the components of
the tool would be substantially different elsewhere. This is a
The CPAx score demonstrated moderate to strong signif- pilot study completed as part of a master’s degree; as such, it
icant positive correlations with GCS score, sedation score was confined by time and finances.
(−3 to 1), muscle strength, AusTOMs score, SF-36 (physical
component) score and peak cough flow. It also demonstrated
a significant negative correlation with SOFA score and the Conclusion
number of days of mechanical ventilation. These results com-
bined suggest validity in the assessment of overall physical This preliminary work demonstrates proof of concept of
morbidity. the CPAx. As a pilot study, it provides a cogent argument
Equally importantly, the CPAx score has been endorsed for investigating this assessment system further in large mul-
by physiotherapists reviewing and using the tool clinically, ticentre studies. Future work should focus on clinician and
and has been shown to have good inter-rater reliability, albeit patient perceptions of the CPAx, further reliability testing,
in a small number of patients. Internal consistency is strong, expert review of the CPAx, and predictive validity for hospital
suggesting that the CPAx can be used consistently with min- outcome.
imal observer error by physiotherapists of varied experience.
However, reliability testing in a larger sample would be ben-
eficial. Acknowledgements
The CPAx assesses (directly or indirectly) all of the com-
ponents of physical morbidity identified as important by a The authors would like to acknowledge the input of those
panel of experts. This tool is unique in its inclusion of respira- involved in the development of the CPAx: Ms. Bronwen Con-
tory domains. In the critically ill, respiratory function and, in nolly, Mr. Gareth Jones, Ms. Alex Curtis; and the support of
particular, weaning from mechanical ventilation is an impor- the clinical teams at the Chelsea and Westminster NHS Foun-
tant component of early rehabilitation [25]. The CPAx allows dation Trust, the Royal London Hospital and St George’s
holistic assessment in the critically ill population, but also University of London.
helps to limit any floor effect, as observed in other tools Ethical approval: Camden and Islington Research Ethics
of this nature (e.g. the Barthel Index). The authors believe Committee (Ref. No. 10/H0722/34).
that the CPAx possesses some of the essential psychometric Conflict of interest: None declared.
properties of a measurement scale.
 E.J. Corner et al. / Physiotherapy 99 (2013) 33–41 39

Appendix 1a. The Chelsea Critical Care Physical Assessment Tool

Aspect of physicality Level 0 Level 1 Level 2 Level 3 Level 4 Level 5

Respiratory function Complete ventilator Ventilator Spontaneously Spontaneously Receiving standard Self-ventilating with
dependence. dependence. breathing with breathing with oxygen therapy (<15 no oxygen therapy
Mandatory breaths Mandatory breaths continuous invasive intermittent invasive l)
only. May be fully with some or non-invasive or non-invasive
sedated/ paralysed spontaneous effort ventilatory support ventilatory support
or continuous high
flow oxygen (>15 l)

Cough Absent cough, may Cough stimulated Weak ineffective Weak, partially Effective cough, Consistent effective
be fully sedated or on deep suctioning voluntary cough, effective voluntary clearing secretions voluntary cough,
paralysed only unable to clear cough, sometimes with airways clearing secretions
independently (e.g. able to clear clearance independently
requires deep secretions (e.g. techniques
suction) requires Yankauer
suctioning)

Moving within the Unable, maybe fully Initiates movement. Initiates movement. Initiates movement. Independent in ≥3 Independent in <3
bed (e.g. rolling) sedated/ paralysed Requires assistance Requires assistance Requires assistance seconds seconds
of two or more of at least one of one person
people (maximal) person (moderate) (minimal)

Supine to sitting on Dynamic Initiates movement. Initiates movement. Initiates movement. Independent in ≥3 Independent in <3
the edge of the bed Unable/unstable Requires assistance Requires assistance Requires assistance seconds seconds
of two or more of at least one of one person
people (maximal) person (moderate) (minimal)

Dynamic sitting (i.e. Unable/unstable Requires assistance Requires assistance Requires assistance Independent with Independent with
when sitting on the of two or more of at least one of one person some dynamic full dynamic sitting
edge of the people (maximal) person (moderate) (minimal) sitting balance (i.e. balance (i.e. able to
bed/unsupported able to alter trunk reach out of base of
sitting) position within base support)
of support)

Standing balance Unable/unstable/be Tilt table or similar Standing hoist or Dependant on Independent without Independent without
dbound similar frame, crutches or aids aids and full
similar dynamic standing
balance (i.e. able to
reach out of base of
support)
Sit to stand (starting Unable/unstable Sit to stand with Sit to stand with Sit to stand with Sit to stand Sit to stand
position: ≥ 90º maximal assistance moderate minimal assistance independently independently
hip flexion) (standing hoist or assistance (e.g. one (e.g. one person) pushing through without upper limb
similar) or two people) arms of the chair involvement.

Transferring from Unable/unstable Full hoist Standing hoist or Pivot transfer (no Stand and step Independent
bed to chair similar stepping) with transfer with mobility transfer without
mobility aid or aid or physical equipment
physical assistance assistance

Stepping Unable/unstable Using a standing Using mobility aids Using mobility aid Using mobility aid or Independent without
hoist or similar and assistance of at and assistance of assistance of one aid
least one person one person person (minimal)
(moderate) (minimal)
Grip strength Unable to assess <20% <40% <60% <80% ≥80%
(predicted mean for
age and gender on
the strongest hand)
40 E.J. Corner et al. / Physiotherapy 99 (2013) 33–41

Age
(years) Men Women
Hand Mean <20% <40% <60% <80% ≥80% Hand Mean <20% <40% <60% <80% ≥80%
15 to R 46.91 9.38 18.76 28.15 37.53 37.53 R 28.82 5.76 11.53 17.29 23.06 23.06
19 L 42.13 8.43 16.85 25.28 33.70 33.70 L 24.98 5.00 9.99 14.99 19.98 19.98
20 to R 48.15 9.63 19.26 28.89 38.52 38.52 R 28.33 5.67 11.33 17.00 22.66 22.66
24 L 43.08 8.62 17.23 25.85 34.46 34.46 L 25.78 5.16 10.31 15.47 20.62 20.62
25 to R 53.76 10.75 21.50 32.26 43.01 43.01 R 33.82 6.76 13.53 20.29 27.06 27.06
29 L 48.60 9.72 19.44 29.16 38.88 38.88 L 30.31 6.06 12.12 18.19 24.25 24.25
30 to R 52.63 10.53 21.05 31.58 42.10 42.10 R 33.97 6.79 13.59 20.38 27.18 27.18
34 L 48.98 9.80 19.59 29.39 39.18 39.18 L 31.64 6.33 12.66 18.98 25.31 25.31
35 to R 53.16 10.63 21.26 31.90 42.53 42.53 R 32.46 6.49 12.98 19.48 25.97 25.97
39 L 51.75 10.35 20.70 31.05 41.40 41.40 L 29.77 5.95 11.91 17.86 23.82 23.82
40 to R 55.49 11.10 22.20 33.29 44.39 44.39 R 30.34 6.07 12.14 18.20 24.27 24.27
44 L 50.40 10.08 20.16 30.24 40.32 40.32 L 26.23 5.25 10.49 15.74 20.98 20.98
45 to R 49.93 9.99 19.97 29.96 39.94 39.94 R 35.30 7.06 14.12 21.18 28.24 28.24
49 L 48.94 9.79 19.58 29.36 39.15 39.15 L 32.06 6.41 12.82 19.24 25.65 25.65
50 to R 48.40 9.68 19.36 29.04 38.72 38.72 R 28.37 5.67 11.35 17.02 22.70 22.70
54 L 41.46 8.29 16.58 24.88 33.17 33.17 L 26.28 5.26 10.51 15.77 21.02 21.02
55 to R 45.71 9.14 18.28 27.43 36.57 36.57 R 29.76 5.95 11.90 17.86 23.81 23.81
59 L 42.16 8.43 16.86 25.30 33.73 33.73 L 27.81 5.56 11.12 16.69 22.25 22.25
60 to R 40.59 8.12 16.24 24.35 32.47 32.47 R 26.35 5.27 10.54 15.81 21.08 21.08
64 L 37.25 7.45 14.90 22.35 29.80 29.80 L 23.47 4.69 9.39 14.08 18.78 18.78
65 to R 40.87 8.17 16.35 24.52 32.70 32.70 R 23.60 4.72 9.44 14.16 18.88 18.88
69 L 36.57 7.31 14.63 21.94 29.26 29.26 L 23.38 4.68 9.35 14.03 18.70 18.70
70 to R 37.48 7.50 14.99 22.49 29.98 29.98 R 25.84 5.17 10.34 15.50 20.67 20.67
74 L 35.49 7.10 14.20 21.29 28.39 28.39 L 22.92 4.58 9.17 13.75 18.34 18.34
R 32.76 6.55 13.10 19.66 26.21 26.21 R 19.40 3.88 7.76 11.64 15.52 15.52
75+ L 28.59 5.72 11.44 17.15 22.87 22.87 L 17.64 3.53 7.06 10.58 14.11 14.11
Table developed from normal UK hand grip strength values in Gilbertson L, Barber-Lomax S. Power and pinch grip strength recorded using the hand-held
Jamar dynamometer and B + L hydraulic pinch gauge: British normative data for adults. Br J Occupat Ther 1994;57:483–8.

Appendix 1b. The Chelsea Critical Care Physical

[4] Hermans G, De Jonge B, Bruyninckx F, Van de Berghe G. Clini-
Assessment Tool cal review: critical illness polyneuropathy and myopathy. Crit Care
2008;12(238), doi:10.1186/cc7100.
Grip Respiratory
[5] National Institute of Clinical Excellence. Rehabilitation after criti-
strength function
5 cal illness. London: NICE; 2009. Available at: www.nice.org.uk, Last
4 Cough accessed: 29/02/2012.
Stepping 3 [6] Collen FM, Wade DT, Robb GF, Bradshaw CM. The Rivermead Mobil-
2
ity Index: a further development of the Rivermead Motor Assessment.
1
Moving Int Disabil Stud 1991;13:50–4.
Transferring 0
within the [7] Fan E, Ciesla ND, Truong AD, Bhoopathu V, Zeger SL, Needham DM.
bed to chair
bed Inter-rater reliability of manual muscle strength testing in ICU survivors
and simulated patients. Intens Care Med 2010;36:1038–43.
[8] Black NA, Jenkinson C, Hayes JA, Young JD, Vella K, Rowan KM,
Sit to et al. Review of outcome measures used in adult critical Care. Crit Care
Stand Supine to sitting Med 2001;29:2119–24.
on the edge of [9] Hayes JA, Black NA, Jenkinson C, Young JD, Rowan KM, Daly K,
the bed
et al. Outcome measures for adult critical care: a systematic review.
Standing Dynamic Sitting Health Technol Assess 2000;4:1–111.
balance
[10] Skinner EH, Berney S, Warrilow S, Denehy L. Development of an
outcome measure (the PFIT): an pilot exercise training protocol for use
in intensive care. Crit Care Resusc 2009;11:110–5.
[11] DiCicco J, Whalen D. University of Rochester Acute Care Evaluation:
development of a new functional outcome measure for the acute care
setting. J Acute Care Phys Ther 2010;1:14–20.
[12] Streiner DL, Norman GR. Health measurement scales: a practical guide
to their development and use. 4th edn. New York: Oxford University
References Press; 2008.
[13] NHS employers. National Profiles for Physiotherapy; 2005.
[1] Dincer EH. How should we deal with muscle weakness in critically ill Last accessed: 28/02/2012 http://www.nhsemployers.org/
patients? Crit Care Med 2009;37:2648–9. PayAndContracts/AgendaForChange/NationalJobProfiles/Documents/
[2] Winkelman C. Bed rest in health and critical illness. AACN Adv Crit Physiotherapy.pdf.
Care 2009;20:254–66. [14] Polit DF, Tatano Beck C, Owen SV. Is the CVI an acceptable indicator
[3] Griffiths R, Hall J. Intensive care unit-acquired weaknesses. Crit Care of content validity? Appraisal and recommendations. Res Nurs Health
Med 2010;38:779–87. 2007;30:459–67.
 E.J. Corner et al. / Physiotherapy 99 (2013) 33–41 41

[15] Lynn MR. Determination and quantification of content validity. Nurs [21] Bigatello LM, Stelfoc HT, Berra L, Schmidt U, Gich I, Gettings E.
Res 1986;35:382–6. Outcome of patients undergoing prolonged mechanical ventilation after
[16] Gao ZJ, Qin YZA. Study of cough peak expiratory flow in predicting critical illness. Crit Care Med 2007;35:2491–7.
extubation outcome. Chin Crit Care Med 2009;21:390–3. [22] Garret AM, Ruta DA, Abdalla MI, Buckingham JK, Russell IT. The
[17] Perry A, Morris M, Unsworth C, Duckett S, Skeat J, Dodd K, et al. SF 36 health survey questionnaire: an outcome measure suitable for
Therapy outcome measures for allied health practitioners in Australia: routine use within the NHS? BMJ 1993;306:1440–4.
the AusTOMs. Int J Qual Health Care 2004;16:285–91. [23] de Rooji SE, Ahbu-Hanna A, Levi M, de Jonge E. Factors that predict
[18] Unsworth C, Duckett S, Duncombe D, Perry A, Skeat J, Tay- outcome of intensive care treatment in very elderly patients: a review.
lor N. Validity of the AusTOM scales: a comparison of the Crit Care 2009;9:R307–14.
AusTOMs and EuroQol-5D. Health Qual Life Outcomes 2004;2, [24] Marik P. Palliative care in ICU: should age limit admission to
doi:10.1186/1477-7525-2-6. the intensive care unit? Am J Hospices Palliat Med 2007;24:
[19] Ali NA, O’Brien JM, Hoffman SP, Philips G, Garland A, Finlay JCW, 63–6.
et al. Acquired weakness, handgrip strength, and mortality in critically [25] Lavine S, Nguyen T, Taylor N, Friscia M, Budak M, Rothenberg P,
ill patients. Am J Resp Crit Care Med 2008;178:261–8. et al. Rapid disuse atrophy of the diaphragm fibers in mechanically
[20] Callahan L, Supinski G. Sepsis-induced myopathy. Crit Care Med ventilated humans. N Engl J Med 2008;358:132–5.
2009;37(Suppl.):S354–67.

Available online at www.sciencedirect.com