Reliability of a new classification system for mobility and self-care in spinal cord injury rehabilitation: the Spinal Cord Injury-Interventions Classification System.

Abstract

To investigate the reliability of the Dutch version of the Spinal Cord Injury-Interventions Classification System.
Descriptive study.
Three Dutch spinal cord injury facilities.
Six physical therapists, 6 occupational therapists, and 3 sports therapists.
Not applicable.
A 4- or 5-digit code was used to score videotaped interventions on 2 occasions in terms of level, category, and type of intervention. We examined the percentage of correctly assigned interventions (agreement between the researcher and the participants), agreement between the first and second measurements (intrarater reliability), and agreement between the therapists of the same discipline (interrater reliability).
A total of 252 interventions were assigned. The agreement between the researcher and the participants was good: the percentage of correct interventions was high at both measurements (91.7% and 94.4%). Intrarater agreement was also good (90.9%). The interrater agreement within the physical therapists and the occupational therapists was high at both occasions (mean, 92.3% and 87.0%, respectively), but lower within the sports therapists (mean, 69.3%).
The study provided first evidence for reliability of the classification system. Further research is needed to confirm these results and to test the reliability of other language versions.

Full text

Page 1

The Spinal Cord Injury-Interventions
Classification System (SCI-ICS)
development and evaluation of a documentation tool to record therapy
to improve mobility and self-care in people with spinal cord injury
door
Sacha van Langeveld

Page 2

Deze studie werd gefinancierd door de Stichting Wetenschappelijk Fonds De
Hoogstraat te Utrecht en het KF Hein fonds te Utrecht.
Het verschijnen van dit proefschrift werd mede mogelijk gemaakt door finan-
ciële steun van de Stichting Wetenschappelijk Fonds De Hoogstraat te Utrecht,
SpinalNet Coloplast B.V. te Amersfoort en Orthopedietechniek De Hoogstraat te
Utrecht.
DeSIgn & Lay-ouT Noenus Design, Soest
Cover IMage & Cover DeSIgn Bart de Boer
prInTeD by Print Partners Ipskamp
ISbn/ean 978-90-393-5358-5
© S. van Langeveld 2010

Page 3

The Spinal Cord Injury-Interventions
Classification System (SCI-ICS)
development and evaluation of a documentation tool to record therapy
to improve mobility and self-care in people with spinal cord injury
De Spinal Cord Injury-Interventions
Classification System (SCI-ICS)
ontwikkeling en evaluatie van een registratie systeem van
paramedische behandelingen gericht op het verbeteren van mobiliteit
en zelfverzorging bij personen een dwarslaesie
(met een samenvatting in het Nederlands)
proefschrift
ter verkrijging van de graad van doctor aan de Universiteit Utrecht
op gezag van de rector magnificus, prof. dr. J.C. Stoof,
ingevolge het besluit van het college voor promoties
in het openbaar te verdedigen op
donderdag 8 juli 2010 des middags te 4.15 uur
door
alexandra Helena bernardina van Langeveld
geboren op 24 november 1962
te ‘s-gravenhage

Page 4

proMoTor:
Prof. dr. E. Lindeman
Co-proMoTor:
Dr. M. W. M. Post

Page 5

Chapter 1
Introduction
7
Chapter 2
Development of a classification system of physical,
occupational, and sports therapy interventions to document
mobility and self-care in spinal cord injury rehabilitation
27
Chapter 3
Feasibility of a classification system for physical therapy,
occupational therapy, and sports therapy interventions for
mobility and self-care in spinal cord injury rehabilitation
45
Chapter 4
Reliability of a new classification system for mobility and
self-care in spinal cord injury rehabilitation:
the Spinal Cord Injury-Interventions Classification System
61
Chapter 5
Contents of physical therapy, occupational therapy, and
sports therapy sessions for patients with a spinal cord injury
in three Dutch rehabilitation centers
81
Chapter 6
Comparing contents of therapy for people with a spinal cord
injury in postacute inpatient rehabilitation in Australia,
Norway, and the Netherlands
103
Chapter 7
General discussion
Summary
Nederlandse samenvatting
129
130
146
150
Dankwoord - Acknowledgements
Curriculum Vitae
List of abbreviations
157
169
173
C
o
n
t
e
n
t
s

Page 6

Page 7

C
h
a
p
t
e
r
1
Introduction

Page 8

8
1. Spinal cord injury rehabilitation
1.1. rehabilitation as a multidisciplinary team effort
Since the second half of the 20th century, there has been great progress in medi-
cal care and rehabilitation for individuals who have sustained a spinal cord
injury (SCI), resulting in improvements in terms of life expectancy, functional
capability and integration into society.1,2 To overcome barriers in participation,
and ‘to add life to years’3, comprehensive management of the wide range of
impairments and activity limitations resulting from SCI requires an interdiscipli-
nary approach by the rehabilitation team.1-3 Rehabilitation is a client-centerd,
educational, problem-solving process that focuses on improving functionality,
ultimately aiming at restoration of social participation and wellbeing.4,5
Rehabilitation begins at the intensive care department5, where treating the first
life-threatening impairments resulting from the lesion (e.g., by decompressing
neural elements, stabilizing the spine and securing tissue perfusion) is crucial to
optimize outcomes. Other essential interventions are those to prevent pressure
ulcers, venous thromboembolism, stress ulceration, bowel impaction, dysphagia
and pulmonary complications. In addition to these primarily medical interven-
tions, careful assessment of neurological impairment contributes to an early
prognosis of possible neurologic recovery and to designing a rehabilitation
plan.5 Early initiation of SCI-specific rehabilitation (e.g., interventions to improve
mobility of the paralyzed limbs, addressing the hand function of patients with
tetraplegia by applying splints) is important, as a delayed start of rehabilitation
interventions may negatively influence a patient’s ultimate functional capability
and increase the length of stay at the rehabilitation center.5-9
After the acute phase, when patients are medically stable, they should preferably
be rehabilitated in specialized units to address their SCI-specific needs and
to optimize medical, functional, social, emotional, recreational and vocational
recovery.6,10,11 The rehabilitation team at such specialized SCI rehabilitation centers
includes a physician; a rehabilitation nurse; physical, occupational, speech,
recreational and vocational therapists; a psychologist and a social worker or
casemanager. Other important professionals, such as rehabilitation engineers or
nutritionists, may be added to the team.1,6 At admission, the physician, who is
the team leader, performs a comprehensive evaluation of the patient’s medical,
neurological and functional status. Based on the findings of this evaluation, the
physician sets short-term and long-term goals and prepares therapy prescrip-
tions for all team members involved in achieving these goals. Guided by the
functional and medical status on admission, the center provides a customized
integrated therapy programme. All disciplines play vital roles in the rehabili-
tation process for the patient and his or her family. Although each discipline
has its own key areas, services do overlap.1,10,12,13 For example, in the training of
Chapter 1

Page 9

9
Introduction
transfers, which is a novel motor task for persons who have suffered an SCI, the
physical therapists focus on practising the transfer skills; they start by teaching
the patient to perform a transfer from a wheelchair to a treatment table, and,
while adapting the pace of progress to the individual patient, they go on to prac-
tice the same task in different situations (e.g., transfer to a car). The occupational
therapists visit patients’ homes to evaluate how they manage transfers at home
and which adaptive equipment and home modifications are needed, while nurses
practice the transfers in the patient’s daily routine. Overlaps between treatment
areas, such as transfers, serve as a reinforcement to patients1,10 ; as repetitious
practice is one of the key features of the effective training of motor tasks.13 Well-
structured team conferences with the patient and key team members are held at
regular intervals to synchronize all therapy activities with the patients’ short-
term and long-term functional goals (e.g., mobility, self-care, domestic life).
Guided by these team conferences, the therapy programme is modified over the
period encompassing the admission phase, the active training phase and the
discharge phase, until the goals have been achieved and the patient is discharged
into the community. In the Netherlands, patients who have been discharged
from inpatient rehabilitation are eligible for an additional period of day-care out-
patient treatment, involving two or more disciplines from the specialized SCI
team. This transitional stage provides them with the opportunity to try out their
new living situation and to evaluate whether the amount and quality of care,
equipment and home modifations is sufficient.14,15
1.2. ICF as the unifying concept of rehabilitation
An important basis for the successful development of rehabilitation practice and
research is a conceptually sound description of rehabilitation, interpreted as a
health strategy based on a universally accepted conceptual model and taxono-
my of human functioning.16 Since the approval by the World Health Assembly
(WHO) of the International Classification of Functioning, Disability and Health
(ICF)17 in 2001 and the reference to the ICF in the WHO resolution16 on ‘Disability,
including prevention, management and rehabilitation’ in 2005, this model has
become universally accepted. (Figure 1)
The ICF17 is a classification of health and health-related states, and is made up of
two parts, each with two components. Part one consists of the components (a)
Body functions and structures and (b) Activities and Participation, while part two
consists of (a) Environmental factors, and (b) Personal factors. Each component,
except Personal factors, consists of various domains, and within each domain,
categories are defined which are the units of the classification.
Body functions are defined as the physiological functions of body systems
(including psychological functions). An example of a body function is muscle
function, which can be specified as power, tone and endurance. Body structures

Page 10

10
are defined as anatomical parts of the body, such as organs and limbs. An exam-
ple of a body structure is ‘muscles of the upper arm’. Impairments are defined as
problems of body functions and structures, such as a significant deviation or
loss. Activity is defined as the execution of a task or action by an individual,
while participation is defined as involvement in a life situation. An example of
the Activities and Participation component is ‘walking and moving’ in the
mobility domain. Activity limitations are difficulties an individual may have in
execu-sting a task, and participation restrictions are problems an individual may
experience in life situations. Although they are presented as different entities in
figure 1, the Activities and Participation categories are offered in ICF17 as one
single list, and it is left to the user to consider a particular category as an activity
limitation or a participation restriction. This lack of clarity has been frequently
criticized in the literature.18,19 The environmental factors make up the physical,
social and attitudinal environment in which people live. The personal factors are
the particular background of an individual’s life and living.17 The environmental
factors and personal factors interact with the components of part one.17
The method used for rehabilitation medicine in the Netherlands has been de-
rived from Bangma (1984)20, and is referred to as the SAMPC method.21 This ap-
proach involves lists of discipline-specific treatment areas (e.g., somatic, activi-
ties of daily living) and communication).21 In 1980, the International Classification
of Impairments, Disabilities and Handicaps (ICIDH)22, the predecessor of the ICF17,
Figure 1. The ICF model (WHo; 2001).17
Health condition
(disorder of disease)
Body Functions
And Structures
Activities
Participation
Personal
Factors
Environmental
Factors
Chapter 1

Page 11

11
Introduction
which classifies consequences of diseases, was widely promoted in many coun-
tries18,19,21, and was translated into Dutch.23 Van Bennekom et al.24 tried to focus
rehabilitation goals on disabilities and handicaps, by developing the Rehabilita-
tion Activities Profile in 1995 (RAP).25 RAP enables users to document restrictions
and limitations in the areas of communication, mobility, self-care and inter-rela-
tional aspects, and evaluates the extent to which these present problems to the
persons involved.26 In 2000, van Dijk21,27 critically evaluated the SAMPC method20,
as well as the ICIDH.22 Inspired by the above mentioned models and the Concept
of Daily Living proposed by Vreede28, van Dijk presented a model of functioning29
which was operationalized for the educational programme used to train reha-
bilitation physicians in the Netherlands. This model29 still remains in use today.
The key concept in this model (referred to as ‘Leerplan’29) is that personal and
environmental characteristics predetermine an individual’s functioning. The
model29 incorporates three levels of functioning: basic functions, basic activities
and goal-oriented activities. The ICF17 was translated into Dutch in 2002.30 All the
concepts described in this paragraph are used to describe the focus of rehabilita-
tion medicine in the Netherlands.
1.3. Differences in SCI rehabilitation in different settings and countries
Worldwide, there are considerable differences in the structures, processes and
outcomes of rehabilitation systems for patients with SCI.31 Structures of care,
i.e., the buildings, materials, expertise and manpower available, constitute the
conditions for delivering care and are relatively stable over time.32,33 Process of
care has been defined as the ‘contents of care, i.e., how the patient was moved
into, through, and out of the health care system’, and the services that were pro-
vided during the care episodes.32,33 The process of care is inherently dynamic and
is the most difficult to measure. Outcomes of care refers to the results of care.32,33
These three characteristics of rehabilitation and their interconnections together
define the quality of rehabilitation care.32,33 For example, rehabilitation centers
may differ in terms of the number and diversity of rehabilitation professionals,
the quality of equipment (e.g., having a technology department to develop cus-
tomized adaptive aids, private wheelchair stock) and facilities (e.g., fully equipped
exercise rooms, swimming pool, indoor sports fields, outdoor activity equip-
ment).31 Differences between rehabilitation centers in terms of these ‘unit charac-
teristics’ might impact on the quality of care.31 Rehabilitation objectives and poli-
cies may also differ: rehabilitation centers might admit patients soon after SCI, or
with a certain delay due to prolonged stay for acute care at a hospital, or they
might be more willing to admit patients with non-traumatic SCI, or might admit
patients of more advanced age.31,34 Different health care systems may have
different regulations for the length of stay (LOS) in inpatient rehabilitation, which
might result in wide variations in LOS.35 Shorter LOS may lead to undesirable

Page 12

12
outcomes (e.g., discharge to nursing homes instead of to the community,
increase in hospital readmissions, and increased long-term costs), but also to
positive outcomes (earlier social reintegration, decreased short-term costs).3,36,37
Although discharge should theoretically occur when a patient’s functional and
educational gains begin to plateau36, some rehabilitation centers may discharge
patients only when optimal functional goals have been achieved and home mod-
ifications have been completed, while others may discharge patients as soon as
it is safe for them to be at home or at a temporary place of stay.34 After discharge,
some rehabilitation centers may offer specialized outpatient care to create a
gradual transition between the safe environment of inpatient rehabilitation and
the ‘real life’ situation in the community, while discharged patients in other
centers have to rely on non-specialized primary care in the community. All these
possible differences influence the contents of rehabilitation and are likely to
have an effect on patient outcomes. Nevertheless, these differences and their
consequences remain largely unknown.31
In the USA, the lack of knowledge about the influence of inpatient SCI rehabilita-
tion on functional outcomes, such as functional independence in terms of
mobility and self-care, has allowed rigorous cost-containment measures to be
implemented. Although these resulted in a considerable reduction of the mean
LOS to 61 days, the effects on rehabilitation outcomes have not been evalu-
ated.3,38 In contrast, LOS of persons with SCI in inpatient rehabilitation in the
Netherlands is much longer, at a mean of 155 days.35 Although to date, the health
insurance system has exerted relatively little pressure to restrict this35,37, there
is an increasing demand for accountability regarding the long duration of
rehabilitation.
2. rationale for the present research project
Knowledge about the relationships between the needs of the patients involved,
the rehabilitation interventions provided and the rehabilitation outcomes
achieved is crucial. Without this knowledge, we cannot identify effective treat-
ments or improve the quality of treatments provided, and it will be difficult
to justify sufficient funding for extensive inpatient rehabilitation as financial
constraints increase.
2.1. research into SCI rehabilitation outcomes
Over the years, rehabilitation research has invested a tremendous amount
of intellectual energy in identifying relevant outcome domains and developing
outcome measures, including psychometric (clinimetric) research on validity,
Chapter 1

Page 13

13
Introduction
reliability, scaling and interpretation of these measures, to provide evidence for
the effectiveness and efficiency of inpatient rehabilitation.38-41 Widely used out-
come measures of SCI rehabilitation include the American Spinal Injury Associa-
tion (ASIA) impairment scale (AIS) to measure neurological recovery,31,42 the
Functional Independence Measure (FIM)41 and the Spinal Cord Independence
Measure (SCIM)43 to measure improvement in daily functioning (e.g., locomo-
tion, transfers, dressing). Length of stay (LOS), defined as the period between
first admission and final discharge from the rehabilitation center, is most com-
monly measured in retrospective studies. In such studies, LOS is used as a proxy
measure of the rehabilitation programme, while functional gain (discharge mi-
nus admission value in scores on the FIM or SCIM) divided by LOS is used as a
measure of rehabilitation efficiency.31 Some authors39,40,44 have characterized and
criticized this as a ‘black box’ strategy of rehabilitation research. Such data were
generated mostly for management purposes, and using them for modelling
outcomes requires cautious interpretation.38,40,45,46 Only Heinemann et al.47, in
1995, and Johnston et al.48, in 2003, investigated the relationship between in-
tensity of therapy and functional status (FIM) at discharge from inpatient SCI
rehabilitation. Their data were obtained in a retrospective study of billing records
and medical reports. They found therapy intensity to be generally unrelated
to any outcome for patients with SCI.47,48 This might suggest that neurologic
recovery (reduced impairment), rather than therapy, may be responsible for the
functional improvement during inpatient rehabilitation.47 Heinemann et al.47 con-
cluded that providing evidence for therapy requires monitoring improvements
over shorter intervals and correlating improvement with the type and extent of
therapy. Since then, no studies have been published in which therapy time or
contents of SCI therapy have been measured directly. Thus, although rehabilita-
tion professionals assume that ‘more of the right kind’ of therapy will result
in better functional outcomes, empirical evidence for a relationship between
the content and intensity of rehabilitation therapy and the functional outcome
remains inadequate.46-48
2.2. research into the contents of SCI rehabilitation
In contrast to the extensive research on SCI outcome measures, little effort has
been invested in research into the actual process of rehabilitation, i.e., the con-
tents of treatments. Although guidelines49,50, books1,10,13,51 and articles2,52,53 giving
valuable guidance on SCI treatments are available, strong scientific evidence for
the effectiveness of SCI rehabilitation interventions is sparse, and much in SCI
rehabilitation is still based on ‘expert opinion’.3,44,52 For example, the recently
published systematic review on physical interventions by Harvey et al.52 found
only six randomized controlled trials (RCT) that reported a clear treatment effect
on at least one outcome measure. These trials supported the use of fitness,

Page 14

14
strength and gait training as well as acupuncture. Topics of recent high level
evidence SCI research include functional electrical stimulation, wheelchair tech-
nologies, partial body-weight-supported ambulation, the use of robotics and the
potential use of brain-control for assistive technology, robots and neuroprosthe-
sis.6 Although the interest in evidence-based medicine has grown, these publica-
tions fail to cover the full spectrum of therapeutic activities provided in SCI
rehabilitation.6
In rehabilitation research, as well as in SCI rehabilitation, it has been stated that
undertaking more RCTs might not be the only way to advance evidence-based
practice.44,54-57 The most important argument for this is that RCTs examine iso-
lated interventions in optimized (standardized) circumstances to detect the
unique contribution of the intervention to the outcome, whereas rehabilitation
involves the simultaneous application of multiple treatments. Interventions are
often not isolated but are reinforced by other interventions by other team mem-
bers. Rehabilitation is an interdisciplinary process that is more than the sum of
its parts, and the call for other practice-based evidence approaches, namely
more rigorous observational studies, has therefore increased.38,44,45,56
2.3. rationale for a classification system
A requirement for both experimental clinical trials and observational research is
to describe the broad range of rehabilitation treatments in objective and opera-
tional terms.39,40,45 A treatment can be described as any intervention that leads to
a sustained change in the natural history or expected course of a patient’s
illness.58 Many authors have emphasized that capturing all interventions in inpa-
tient rehabilitation requires a classification system (also known as taxonomy), in
which interventions are grouped on the basis of similarity into categories that
are objective and mutually exclusive.39,40,44,45,54,59 A classification system of inter-
ventions can guide us in comparing therapy provided during inpatient rehabil-
itation between patients or subgroups of patients (e.g., tetraplegia, paraplegia,
traumatic or non-traumatic, younger and older patients), in identifying the inter-
ventions having the greatest impact on outcomes, and in improving effective
and efficient clinical decision-making based on more than just ‘expert opin-
ion’.39,44 Since no such classification system of treatments for patients with a SCI
was available at the start of the present research project, we decided to develop
a classification of interventions which could describe and compare therapy
across the continuum of SCI care (acute, post-acute and chronic phases), and
which would be site-neutral. In our study we used the term classification system
in the sense of the ordering of entities into groups on the basis of similarity.59
Chapter 1

Page 15

15
Introduction
3. Classifications available at the start of the study
There is a recognized need to address the knowledge gap regarding the contents
of therapy in stroke rehabilitation, a patient population representing the most
commonly treated group in acute and subacute rehabilitation settings.33,38-40,45,60
Consequently, classification systems have been developed and used to record
therapy, in terms of time and contents. These existing systems have influenced
the development of our own classification system.39,60-63 The publication by
De Jong et al.39 was the most important paper to serve as an example for our
classification. In this publication, the authors describe two approaches to devel-
op a taxonomy of rehabilitation interventions, and present the physical therapy
taxonomy. Their occupational therapy taxonomy was published later, and could
therefore not be used as a model in the development of our classification.64
The physical therapy taxonomy by De Jong et al. identified 10 so-called ‘target
functional activities’ (pre-functional activity, bed mobility, sitting, transfers, sit-
to-stand, wheelchair mobility, pre-gait, gait, advanced gait, community mobility),
and two additional non-function-related activities, as the core structuring
feature. As a second dimension, it identified a list of ‘body systems’ with four
subcategories (neuromuscular, musculoskeletal, cardiopulmonary and cogni-
tive-perceptual-sensory interventions). We argued that the use of their system
was limited by their decision to label these functional activities, using terminol-
ogy (e.g., ‘pre-gait’) that had been agreed upon by clinical experts from a limited
number of specialized SCI centers, but not globally accepted; what if therapists
outside these centers did not agree with the terminology used? Also, the inter-
ventions listed in the subcategories of body systems were very diverse, not
meeting the basic requirements for a taxonomy in which activities should be
grouped by similarity. For example, their musculoskeletal interventions subcate-
gory covers (1) balance training, (2) postural awareness, (3) motor training, (4)
proprioceptive neuromuscular facilitation (PNF), (5) neuro-developmental treat-
ment (NDT), (6) gait training using body weight support, (7) upper extremity in-
volved and (8) constrained induced movement therapy. NDT and PNF are spe-
cific methods as such, while balance training is the aim of a treatment and can
be applied using different methods. And the involvement of a particular ex-
tremity belongs to yet another dimension. We argued that this diversity in one
subcategory would create confusion, and possible misclassification (what if a
therapist used a different method to improve muscle power than PNF?; what if
therapists used gait training in water?
This issue was also recognized as posing a risk of misclassification in the
other studies on taxonomies.60-63 Some of the more generic descriptors from the
physical therapy taxonomy, such as education, equipment, modality interven-
tions, assessment, and group and individual therapy, proved valuable for the