Introduction

The effect of the continuous improvement in care of people with spinal cord injury (SCI) is reflected by the increased survival in the past 20 years.1 Ideally, people after an SCI are treated in specialized facilities providing interdisciplinary team care. Teams consisting of physicians, nurses, psychologists, social workers, physical and occupational therapists and other health professions meet the large variety of clinical needs of acute and post-acute SCI patients, resulting in more and faster functional improvement. The provision of comprehensive follow-up care enables SCI individuals to improve and sustain their achieved level of functioning and prevent complications. Assessment tools based on a shared terminology and common definitions are needed to identify the needs of SCI individuals, and to assess the effects of interventions on their health status. However, a common language requires an etiologically neutral framework describing comprehensively the functioning and health of the individuals and one that is accepted by all health professions involved.

Since the approval of the International Classification of Functioning, Disability and Health (ICF) by the World Health Assembly in May 2001, such a common framework and language for all health professionals exists.2 The ICF is based on the integrative model of functioning and comprises four components—Body Functions, Body Structures, Activities and Participation, and Environmental Factors. Realizing that the significance and power of the ICF lies in its conceptualization of functioning and disability, there is an urgent call for creating ICF-based tools that are more appropriate for the clinical practice. The ICF Core Sets are corresponding to this need. ICF Core Sets are lists of categories that are relevant to persons with a specific condition or in a specific setting.3

When defining ICF Core Sets, a number of perspectives, including those of the patients and experts, should be considered and integrated in the development process.3 One method of capturing the experts’ view is to perform a worldwide survey including health professionals experienced in the treatment of people with SCI to identify the problems in functioning.4 The overall objective of this study is to explore the expert perspective on the problems of individuals with SCI in the early post-acute and long-term context. The specific aims of the study were (1) to identify problems of individuals with SCI in the early post-acute and the long-term context, respectively, addressed by health professionals and (2) to summarize these problems using the ICF.

Materials and methods

Study design

The study was conducted as a worldwide Internet survey including physicians, nurses, physical therapists, occupational therapists, social workers and psychologists.

Recruitment procedure and study population

The starting point of the recruitment of participants was a list of facilities specialized in the treatment of SCI. A contact person of each facility was identified through an Internet search. These people received an e-mail including detailed information about the expert survey and the request to nominate health professionals working at their facility and who would be interested to participate in the survey. Health professionals had to meet the following criteria to be included as ‘experts’ in the survey: (1) the professional background had to be a nurse, social worker, physical therapist, occupational therapist, psychologist or physician with different specializations; (2) experience in the treatment of adult individuals with SCI; and (3) fluency in English language to contribute to the survey. All health professionals meeting these inclusion criteria made up the so-called ‘expert pool’. In the next step, we drew a random sample out of the expert pool who were stratified by professional background and World Health Organization regions (http://www.who.int/about/regions/en/index.html) to assure that the different professional perspectives as well as the regional perspectives were taken into account. An exception was made for the ‘African Region’ and the ‘Eastern Mediterranean Region’; all nominated experts of these regions were included as the number was extremely small.

According to a power of 0.8 and a level of significance of 0.05, a sample size of 204 experts would be necessary to determine frequencies with a precision of 10%. Under the assumption that the dropout rate would be approximately 20%, the final study sample consisted of 243 experts.

Measure

The questionnaire applied in the expert survey consisted of two parts. In part I, we asked for basic information about the participant and the professional background. This minimum information was required to get a clearer picture of our study population. In part II of the questionnaire, we asked the participants to list the problems in the functioning, as well as relevant environmental and personal factors of individuals with SCI. The open-ended questions are shown in Table 1. The participants were asked to differentiate between problems experienced in the early post-acute context and in the long-term context. The borderlines for the ‘early post-acute context’ and for the ‘long-term context’ applied in the preliminary studies of the project ‘ICF Core Set development for Spinal Cord Injury (SCI)’ were defined as follows: The early post-acute context begins with active rehabilitation and ends with the completion of the first comprehensive rehabilitation after the acute SCI. The long-term context follows the early post-acute context. This definition was applicable throughout the world, irrespective of the different health systems.

Table 1 Questions applied in the expert survey
Full size table

Data collection procedure

All selected experts got an e-mail with detailed information about the expert survey and the further procedure. The second e-mail included the detailed instructions, the user name and the password to get access to the website created for the Internet survey. In case the experts experienced any technical problems when entering data on this website, we provided a word document based on the same form as used in the Internet survey. Reminders were sent after 1, 2 and 3 months. All answers were kept anonymous.

All answers retrieved from the experts were translated (‘linked’) to the ICF based on established linking rules.5 The ICF includes more than 1400 disjunctive categories, which are structured in a hierarchical manner. The letters b, s, d and e refer to the components Body Functions, Body Structures, Activities and Participation, and Environmental Factors, respectively. They are followed by a numeric code starting with the chapter number (one digit), for example, b4 Function of the cardiovascular, hematological, immunological and respiratory system, followed by the second level (two digits), for example, b410 Heart functions, and the third and fourth levels (one digit each), for example, b4101 Heart rhythm.

An answer could be linked to one or more ICF categories, depending on the number of themes contained in the answer. If an answer was too general to allow a decision on the linkage to a specific ICF category, the concept was regarded as ‘not defined’. For example, concepts such as ‘adapted physical environment’ or ‘inadequate living situation’ were regarded as ‘not defined’. If an answer pertained to personal factors that are not coded within the system of the ICF, the code ‘personal factor’ was attributed. To give an example, concepts such as ‘acceptance of condition and changes to lifestyle’ or ‘coping strategies for being disabled’ were regarded as ‘personal factors’. If an answer describes an aspect of functioning and health that is not covered by the ICF, the code ‘not covered’ was attributed. To give an example, ‘loss of privacy’ was regarded as ‘not covered’. If an answer pertained to health conditions that are not coded within the system of the ICF, the code ‘health condition’ was attributed. To give an example, concepts such as ‘deep vein thrombosis’ or ‘fractures’ were regarded as ‘health conditions’.

Quality assurance procedures

All researchers involved in the linkage were thoroughly trained in a 1-week workshop performed by team members of the ICF Research Branch Munich. During the linking exercise, MS supervised all researchers. After the linking of all answers was completed, MS checked the results regarding irregularities and implausibility. To assure and test the accuracy and rigor of the linking process, 4.2% of the answers were randomly selected and linked by a second researcher of the ICF Research Branch.

Data analysis

The absolute frequencies and relative frequencies of the linked ICF categories are reported, stratified by the context along with their 95% confidence intervals. ICF categories are presented at the second level. If a concept was linked to a third- or fourth-level ICF category, the corresponding second-level category is reported. For example, the corresponding second-level ICF category for the third-level category b4101 Heart rhythm is b410 Heart functions. This is appropriate because the lower-level categories share the attributes of the higher-level category.2 If an ICF category was assigned repeatedly to the answers of one participant, it was counted only once to avoid bias. In this study, a cutoff point of 5% was chosen for the frequencies of the linked ICF categories. This means that all second-level ICF categories linked to concepts reported by at least 5% of the experts are reported. A list of ICF categories on all levels of the ICF can be retrieved from the authors.

The reliability of the linkage process was evaluated by calculating the κ-coefficient and nonparametric bootstrapped confidence intervals.6 The values of the κ-coefficient range from 0 to 1, in which 1 indicates perfect agreement and 0 indicates no additional agreement beyond what is expected by chance alone. The κ-analysis was performed using SAS.7

Results

Recruitment and participants

A total of 211 facilities that were specialized in the treatment of SCI were contacted worldwide. Totally, the expert pool comprised 467 health professionals from 63 facilities. The expert survey was conducted between August and October 2006. Out of 243 selected experts, 144 (59.3%) filled in the questionnaire. However, 22 experts could not be included in the data analysis as they had not completed the questionnaire (n=19) or they were specialized in the treatment of adolescents with SCI (n=3).

The participants were 24 to 62 years old (median 40.0 years); 75 (61.5%) participants were female. Most of the participants were physical therapists (27.9%), followed by physicians (21.3%), occupational therapists (18.9%), nurses (13.9%), social workers and psychologists (both 9.0%). Among the participants, 41.0% were from the European region, 27.0% from the region of the Americas, 14.8% from the Western Pacific region, 9.8% from the Southeast Asian region, 4.9% from the African region and 2.5% from the Eastern Mediterranean region. The main working field of the participants was ‘clinical practice’ (86.1%), followed by ‘research’, ‘management’ and ‘education’. The mean number of years of general professional experience was 15.8 years (s.d. 9.3). The mean number of years of professional experience with SCI patients was 12.1 years (s.d. 8.8).

Expert survey

In total, 7650 different themes (‘concepts’) could be retrieved from the answers of the experts. Of these concepts, 78.8% could be linked to ICF categories: 114 concepts were linked to the ICF components, 601 to first-level ICF categories, 3.471 to second-level ICF categories, 1.726 to third-level ICF categories and 119 to fourth-level ICF categories. In all, 6.6% of the concepts were regarded as ‘not defined’, 8.5% were regarded as ‘personal factors’, 4.0% were regarded as ‘health condition’ and 2.0% were not covered by the ICF.

Tables 2, 3, 4 and 5 list the second-level ICF categories that were linked to concepts reported by at least 5% of the experts. The results are shown stratified by the health-care context. In the early post-acute context, 27 of the 88 categories (30.7%) belong to the component Body Functions, 13 (14.8%) to the component Body Structures, 27 (30.7%) to the component Activities and Participation, and 21 (23.9%) to the component Environmental Factors. A total of 16 ICF categories were unique for the early post-acute context; that is, these categories were identified as relevant for only this context. In the long-term context, 25 of the 92 categories (27.2%) belong to the component Body Functions, 12 (13.0%) to the component Body Structures, 33 (35.9%) to the component Activities and Participation and 22 (23.9%) to the component Environmental Factors. In all, 20 of these ICF categories were unique for the long-term context.

Table 2 Relative frequency of second-level categories of the ICF linked to the themes contained in the answers of the 122 participants: component Body Functions
Full size table
Table 3 Relative frequency of second-level categories of the ICF linked to the themes contained in the answers of the 122 participants: component Body Structures
Full size table
Table 4 Relative frequency of second-level categories of the ICF linked to the themes contained in the answers of the 122 participants: component Activities and Participation
Full size table
Table 5 Relative frequency of second-level categories of the ICF linked to the themes contained in the answers of the 122 participants: component Environmental Factors
Full size table

The κ-coefficient for the linking was 0.57, with a 95% bootstrapped confidence interval of 0.50–0.63.

Discussion

To our knowledge, this is the first Internet survey examining the problems of individuals with SCI from a worldwide perspective, involving six different health professions. The ICF was used as reference as it provides a neutral framework and language. A large variety of aspects in functioning and health were identified by the experts, which are common for both the early post-acute and the long-term context. Nevertheless, unique aspects for each of the two contexts could be identified.

The number of relevant ICF categories for individuals with SCI in the early post-acute context and in the long-term context was 88 and 92, respectively. The majority of the retrieved ICF categories refer to impairments in body functions and restrictions in activities and participation. These results reflect the fact that SCI is multifaceted as it may involve all body functions below the level of the neurological lesion.

All chapters of the ICF component Body Functions were represented in the answers of the health professionals, with the exception of Chapter 3, Voice and speech functions. The most frequently named body function was b280 Sensation of pain. The relevance of this finding is supported by several studies showing that most people with SCI experience chronic pain with high pain intensity over multiple body locations.8 Other frequently named categories refer to functions of respiration, defecation and urination, which are well-known problems associated with SCI.9, 10, 11 Regarding neuromusculoskeletal functions, the most frequently linked ICF categories, b735 Muscle tone functions, b730 Muscle power functions and b760 Control of voluntary movement functions, address problems such as spasticity and coordination of movements.12 Although the categories b152 Emotional functions and b420 Blood pressure functions were more prominent in the early post-acute context, the category b640 Sexual functions was more frequently identified for long-term SCI people. It is reported that many newly injured people who met the diagnostic criteria for depression remit within 3 months of onset,13 as well as that orthostatic hypotension is particularly evident in the early phase.14 The relevance of sexual functions is in line with a survey examining the determinants of quality of life in people with SCI, showing that paraplegics assigned the highest priority to the regain of sexual function.15

It is not surprising that the most frequently named body structures correspond to the body functions listed above, such as structures related to movement, structure of intestine and structure of urinary system. The ICF category, s810 Structure of areas of skin, addresses the consequences of impaired protective functions of the skin and reflects the relevance of pressure sores in these patients.12

Clear differences between the early post-acute and the long-term context came up regarding the ICF component Activities and Participation. ICF categories assigned to Chapter 6 Domestic Life, as well as categories addressing intimate relationships, employment and recreation/leisure were more frequently named for the long-term context. Nevertheless, a variety of elements of Chapter 5 Self Care, were frequently identified for individuals in both contexts, reflecting the fact that their restrictions in daily life are permanent. This result is supported by the study of Liem16 reporting a 42% increased odds of needing more help with activities of daily living per decade after SCI.

Accordingly, the experts considered the support provided by family members, friends and health professionals as a relevant aspect throughout the continuum of SCI care. When planning the transition from the facility to the community, this environmental factor might become a key element.17 Some environmental factors that are related to community participation and labor, such as e150 Design, construction and building products and technology of building for public use and e590 Labor and employment services, systems and policies, were more frequently named as being facilitators and/or hindrances for people with SCI in the long-term context. Health services, systems and policies proved to be the most frequently named environmental factor for both contexts. This finding is supported by Whiteneck et al.,18 who analyzed the importance of environmental factors from the patient's and reported that the provision of health care was one of the main environmental barriers.

It stands out that in general a relatively low proportion of the health professionals named the same problems. It remains an open question whether this proportion would have been higher when providing them with a list of problems requesting to mark the important ones for SCI.

In general, the Internet-based survey proved to be an appropriate method for this study objective. The response rate of 59% was comparable with other studies.4 Experts from all of the six world regions as defined by the World Health Organization could be recruited, guaranteeing a wide range of expert opinions. However, specifically the African and the Eastern Mediterranean region seem to be underrepresented. This may be associated with a restricted access to e-mail and Internet technology for health professionals and the limited knowledge of English in these countries on the one hand. On the other hand, this may also reflect a potential lack of SCI experts within a specific health profession in this region. The procedure of linking the patients’ problems reported by health professionals to the appropriate ICF categories proved to be complex and not at all simple. The κ-coefficient of 0.59 reached in this study reflects a ‘moderate’ agreement between the two people who performed the linking. It is slightly lower than in other studies that used the same linking method.19, 20 It is essential to mention that the exploration of the view of health professionals is insufficient when studying the functional problems of people with SCI. To identify all relevant ICF categories for SCI, the expert perspective reported in this paper was complemented by studies exploring the perspectives of the patients, the clinics and the researchers.3 The results will be published elsewhere.

Conclusion

Health professionals address a large variety of functional problems related to SCI that reflect the complexity of this condition. Unique aspects of functioning exist for the early post-acute context and the long-term context, respectively. The ICF proved to be a neutral framework and language, allowing the comparison of answers collected from different professional backgrounds and from different world regions.