Introduction

Work disability (WD) may occur as a result of a health condition or a health-related event and the associated burden may be evident at the individual level and the society level. This burden may appear in the form of consequences of limited or restricted work participation. Vocational rehab (VR) is a key process in WD management which aims to engage or re-engage individuals back to work and employment and has been documented in the literature to be effective in addressing WD issues [1–6].

The International Classification of Functioning, Disability and Health [7] is a generic conceptual framework and classification system of the World Health Organization (WHO) that can be actively interfaced with VR or within the context of return-to-work, or work participation [8–11]. As a conceptual model, the ICF recognizes that functioning and disability is a result of the interaction between components: body functions (b), body structures (s), activities and participation (d), environmental factors (e), and personal factors (not coded). As a classification system, the ICF can serve as a basis for evaluating the scope and complexity of VR services by providing a comprehensive list of functioning domains in the form of alphanumeric coded ICF categories that are arranged in a hierarchical fashion, hence different levels, for each of the ICF categories or functioning domains. Below is an illustration of this categorization:

 

ICF component

d activities and participation

Chapter

d4 Mobility

Second-level category

d430 Lifting and carrying objects

Third-level category

d4300 Lifting

However, there remains a lack of understanding of functioning within VR in terms of the ICF, despite the need for VR providers to understand the broad range of functioning factors (instead of being too focused) to improve work disability outcomes in the clinical setting or community-based setting in VR. Therefore, the objective of this study is to survey the experts in the VR field regarding what factors are important to individuals participating in VR using the ICF as the language to summarize the results.

Materials and Method

Recruitment of Experts

A survey was conducted over the internet with expert participants from six WHO Regions (Africa, the Americas, Eastern Mediterranean, Europe, South-East Asia, and Western Pacific). An “expert” was arbitrarily defined to have at least 2 years of relevant experience (practice, research, or both) in the field of VR, may be a health professional, and is able to communicate competently in English [12]. A procedure using Secure Sockets Layer, 128-bit encryption technology ensured the security and privacy of the web-based survey.

Invitations to the experts were sent out via secure electronic e-mail. E-mail addresses of experts were obtained from literature search (through a concurrent systematic review), internet search, professional organizations, journal editorial board, informal networks, and peers. An expert who was contacted was also asked to name and refer other experts.

Recruitment of experts into the “pool” was performed from March to June 2009. Selection of experts from this pool who were to complete the survey was done using randomized stratified sampling. This sampling method was modified so that at least one expert from each country is included. This modification was made to ensure that all the countries in the pool are represented.

Information from the Experts

Sociodemographic information was collected from the sampled experts. Information included country, age, sex, profession, main field of practice (>50% of the time), years of experience in VR, whether they are a direct healthcare provider, primary type of patients they deal with, and types of VR services they are involved with. Experts were also asked to self-rate based on an 11-point numerical scale regarding their expertise in VR (from 0 = “No experience” to 10 = “Excellent experience”).

Survey Website and Survey Questions

The survey was conducted in English. The initial part of the survey asked the respondents information on sociodemographics and their experience in VR. For the main part of the survey, experts were asked with six open-ended questions (see Table 1) to examine which factors they consider relevant and important to individuals who participate in a VR program. Each question is related to a component of the ICF (body functions, body structures, activities and participation, environmental factors, and personal factors). For example, the first question on “body and mind” is related to the ICF component of body functions. Experts, however, did not see the ICF component labels embedded in the questions. Blank fields for the answers were provided after each question and respondents were allowed to provide multiple answers. Answers were not limited in terms of word length although respondents were instructed to be brief and concise and avoid abbreviations and vague technical terms. The expected completion time for the survey was 30 min.

Table 1 ICF-based questions that the experts were asked. The ICF component indicated within the brackets were not seen by the participants
Full size table

Linking to the ICF

All responses gathered from the experts were linked to the ICF applying published linking rules [13]. The objective of the linking process is to be able to translate the concepts found in the experts’ responses into the most fitting ICF categories. Two individuals who are trained and experienced in the linking process, independently linked the responses. The first linker (L1) linked 100% of the responses, while the second linker (L2) counter-linked 33% randomly selected concepts of L1. If there was a disagreement between L1 and L2 with the counter-linked concepts, both tried to resolve the disagreement between them, otherwise a third person was consulted.

The ICF categories that were identified based on the experts’ responses were listed and a frequency analysis was performed.

Reliability of Linking

To evaluate the reliability of the linking process, the overall percentage of agreement was calculated based on the two independent linkage versions. In addition, to examine the extent to which the achieved agreement exceeds chance, the Kappa coefficient [14] and nonparametric bootstrapped confidence interval [15, 16] were calculated. We performed Kappa analysis using SAS software version 9.1 (Copyright © 2002–2003 by SAS Institute Inc., Cary, NC, USA).

Results

Recruitment of Experts

There were 626 experts recruited into the pool who were identified from our search and were invited to have their names in the pool. Out of the 626 experts, 295 (47%) from 47 countries agreed to be in the pool. After a modified stratified randomized sampling (profession × WHO region × country), 201 experts were selected to whom the survey was sent. In total, we received responses from 151 experts (75% response rate). Electronic reminders (i.e. e-mail) were sent to the experts twice in regular interval to encourage participation.

Characteristics of Expert Respondents

Sociodemographics of the respondents and their self-rating of expertise are presented in Table 2. About one-third of the respondents belonged to the 41–50 year-old age range; nearly half of them were male, a little over a third were from Europe, close to two-thirds of the sample were physical and occupational therapists, and most respondents were engaged in clinical or research work. A majority of the respondents (>85%) had at least 8 years of experience in the field of VR and have rated themselves at least 7 out of 10 in the “expertise scale” of 0–10 with number close to 10 meaning greater expertise. About half of the respondents directly provide healthcare service to patients and close to 65% conducts their work on patients with musculoskeletal health conditions. Looking at the specific types of VR services that the respondents provide or are engaged with in research, we found a broad variety of those services (Table 2).

Table 2 Demographics and VR experience of experts surveyed (percentage of n = 151 expert respondents)
Full size table

Description of Responses and ICF Categories

Based on the responses generated from the survey, most of the respondents provided topical words and phrases. Mean time spent to complete the survey for 128 respondents was 29 min and 30 s (median = 21 min and 22 s). Completion time for the rest of the experts was excluded because they either took multiple days or longer than 8 h to complete the survey. The survey platform was configured to allow them to answer parts of the survey in different times. This means that a respondent may start answering a few questions today and then complete the survey tomorrow, so that will indicate more than 24 h of survey completion when in reality it was probably just 30 min in total.

One hundred one second-level ICF categories were identified based on the ICF-linked responses of the experts. Only those that were stated by at least 5% (arbitrary decision) of the respondents were included. Out of these 101 categories, 22 (21.8%) were related to body functions (Table 3), 13 (12.9%) were related to body structures (Table 4), 36 (35.6%) were related to activities and participation (Table 5), and 30 (29.7%) were related to environmental factors (Table 6).

Table 3 Body functions (N = 22) (included were only those categories mentioned by at least 5% of the respondents)
Full size table
Table 4 Body structures (N = 13) (included were only those categories mentioned by at least 5% of the respondents)
Full size table
Table 5 Activities and participation (N = 36) (included were only those categories mentioned by at least 5% of the respondents)
Full size table
Table 6 Environmental factors (N = 30) (included were only those categories mentioned by at least 5% of the respondents)
Full size table

Our reliability calculation between linkers (researchers) at the second-level ICF categories resulted in an overall percentage agreement of 61.7%, an estimated 0.61 Kappa coefficient, and a confidence interval (bias corrected percentile method) of 0.58–0.63. For the third-level categories, the overall percentage agreement was 56.0%, estimated 0.55 Kappa coefficient, and a confidence interval (bias corrected percentile method) of 0.53–0.58.

Discussion

VR is a key process that enables individuals to participate in or return back to gainful employment. VR, as a component of work disability management, covers a wide variety of factors. In this study, we used the ICF as a conceptual framework and classification system to explore and examine factors surrounding VR and this study benefits from the ICF language and its cross-setting application. We conducted a worldwide survey of experts who are involved in VR. We asked questions on what are the important factors that need to be considered in VR and linked the factors to the ICF so it can be meaningful to ICF users. We found that there is a multitude of ICF functioning domains based on the experts’ responses which indicates further the complexity and breadth of VR as a field of research and practice.

The experts’ perspective has been documented and reported in the literature [17–21]. Information gathered from the experts in the area provides a unique and rich source of knowledge to understand workers with disability. This knowledge will further be useful for health care providers and clinical researchers.

The experts who have participated in this survey had different backgrounds. They came from 47 different countries, a fact which contributed to the multicultural and multinational perspective. This characteristic would also have an important implication in terms of possible applicability of this study to other settings. While it is remarkable that experts from all six WHO regions participated in this study, it is interesting to note that most of them came from developed regions such as Europe (primarily Western Europe), the Americas (primarily USA and Canada), and Western Pacific (primarily Australia). There is a great imbalance of the number experts from the different regions (only 12 out of the 151 experts, were from the WHO regions of Africa, Eastern Mediterranean, and Southeast Asia). The practice of VR could be more established in developed nations where VR infrastructure is in place, relative to developing or undeveloped nations. Indeed, VR services and its way of delivery vary even among developed countries [22, 23] due to difference in system and politics. The variation between world regions could be associated with variation in the availability of established health, economic, labor, and social systems that may be able to provide vocational rehabilitation services.

The respondents had different professions bringing their own unique and shared experience either into a unidisciplinary or interdisciplinary VR setting. This finding gives an indication of broad scope of VR given the different disciplines to address return to work or increasing work participation. Most of the respondents were therapists (physical or occupational therapist). However, we learned that different professionals from some countries have different scope of practice in providing VR. In South Africa, physical therapists usually do not see patients for VR but occupational therapists do. A high proportion of respondents represented the category “other professions”. Looking closely at this category provides an insight as to how broad VR practice and research areas are—from movement science to sociology, public health, and health economics. Since our recruitment for experts was inclusive rather than exclusive, we were not surprised to find a variety of different fields of VR practice—from experts who directly manage patients to those engaged in clinical trials, teaching, and management. Given the multidisciplinary practice of VR, it is likely that experts do cross-over from one setting to the other within the same period (e.g. treating patients and teaching at a university).

The level of experience of the respondents was remarkable—the majority of them have 8 years of experience at a minimum, which gives credible weight to their input on the study. Moreover, a majority of the respondents rated themselves high in terms of their expertise in the field of VR. Our sample represented a good variety of experts who provide direct care to patients and those who work in research or administration—two perspectives that are essential.

The health conditions most commonly treated or researched by our respondents were musculoskeletal, mental, and neurologic health conditions. This finding is indicative of the high prevalence and great socioeconomic burden that these health conditions (alone or in coexistence) pose on patient level [24–27].

The diverse characteristics of the respondents were also evident in terms of the VR services or programs they are engaged with. Although the list of VR services was not exhaustive, it gave us an indication of the complexity and breadth of VR. Data on “other VR services” was collected to ensure that the full spectrum of services possible was covered which provided us additional information on VR services such as consultation, caregiver training and work and home management.

The experts’ perspective based on this survey covered all four classifiable components of the ICF: body functions, body structures, activities and participation, and environmental factors. Of these components, activities and participation represented the most categories followed by environmental factors and body functions. The least representation could be found by body structures categories. This finding signifies a broad societal perspective rather than a sole consideration for the individual structure-level alone. This broad representation of the ICF components was found to be essential if VR or a return-to-work were to be successful [28–31].

The ICF categories selected based on the expert survey is reflective of VR as a multifaceted and multifactorial process. Work resumption as a function of work status is a common indicator of VR success [3, 32–34]. In the component activities and participation, therefore, it was not unsurprising to find remunerative or gainful employment (d850 remunerative employment) being the most frequently mentioned category. Other work-relevant categories were d845 acquiring, keeping and terminating a job, d240 handling stress and other psychological demands (perhaps applicable to mental-type of health conditions or jobs that demand mental competency), and d440 fine hand use and d430 lifting and carrying objects (perhaps applicable to physical-type of health conditions or jobs that demand physical skills). On a different note, non-remunerative (non-paid) employment categories (d855 non-remunerative employment) such as volunteering were considered important along with d920 recreation and leisure, d640 doing housework, d230 carrying out daily routine which are not necessarily traditional work domains. This evidence on work and non-work factors relevant to VR is essential in understanding work disability in general and in the delivery of successful VR [31]. Further, it supports the notion that “employment” or work does not necessarily have to be paid to be considered as such.

With regard to categories from the component environmental factors, it was evident that support from people surrounding the worker is essential. This support may come from family members (e310 immediate family, e315 extended family), boss or employer (e330 people in positions of authority), and co-workers (e325 acquaintances, peers, colleagues, neighbours and community members). This finding is consistent with the role that support and relationship play as part of one’s social environment [8, 34]. The physical and at times political environment does play a crucial role in facilitating work participation [35–37]. Support from health care services and providers (e580 health services, systems and policies and e355 health professionals), labour services and providers (e590 labour and employment services, systems and policies), and social security services and providers (e570 social security services, systems and policies) was also considered by the respondents to be essential. These services related to health, labour, and social security appear to be the “trifecta” in facilitating work participation. Within the context of remuneration, assets (e165) in form of money, income, salary, or benefits could indeed facilitate or sustain return to work [8]. Other physical infrastructures were also observed such as transportation (e540 transportation services, systems and policies), building accessibility (e150 design, construction and building products and technology for public use and e155 design, construction and building products and technology for private use), and the workplace (e135 products and technology for employment).

The categories under the body functions component of the ICF lend its applicability to different kinds of health conditions and types of work, either physical or mental, or a combination of both. Mental health-related categories that were evident from the experts’ responses include b126 temperament and personality functions, b164 higher-level cognitive functions, b117 intellectual functions, and b140 attention functions as examples. Physical health condition-related categories included b730 muscle power functions, b760 control of voluntary movement functions, and b710 mobility of joint functions, as examples. On one hand, there might be VR factors that are relevant across health conditions but may have varying effects or influence depending on whether it is mental or musculoskeletal condition in nature [34]. On the other hand, in some VR participants with certain health conditions and job types, both mental and physical-relevant categories may be applicable.

The respondents also included several categories from the body structures component that were important in VR. A few items but a comprehensive list included structures that were relevant to mental, musculoskeletal, and neurologic health conditions that appeared to be consistent with type of health conditions the respondents commonly deal with. These structures included brain (s110 structure of brain), back and extremities (s730 structure of upper extremity, s750 structure of lower extremity, and s760 structure of trunk). Also included were structures of special senses such as the eye (s220 structure of eyeball) and ear (s250 structure of middle ear and s260 structure of inner ear). This coverage reflects the broad spectrum of diseases in VR settings that experts encounter.

We did not make a separate analysis for inter-country or—region difference of responses. We assumed that the ICF functioning domains operate regardless of country, region, or setting. We suspect, however, that there might be variation at the granular level of the systems, politics, governance, social environment, and services—which overall may still be similar in ICF terms but whose operationalization may be different. Nevertheless, the “conceptualization” of the different domains in different countries may not be different, in our opinion.

With regards to our linking methodology, there was not only a satisfactory agreement between the linkers but the agreement also exceeded chance. This shows that the linking procedure was reliable in this study.

We do recognize the limitations of this study. First, the results could not be generalizable to all experts’ perspectives on VR. While there was high response rate, it is possible that some experts who have been contacted did not receive the invitation due to incorrect e-mail addresses, or that experts from some developing nations do not have access to e-mail technology hence, were not included in the initial pool to begin with. Further, only those experts who were competent in the English language (self-reported) were included which could lead to selection bias. Secondly, the responses to the survey were not verified for possible misclassification because they were all self-reported. Thirdly, different countries have different means of VR so for example, only physical therapists are allowed to provide VR or VR services are provided for physical health conditions only and not for mental conditions. Fourthly, the type of VR services that the experts were involved with were not exhaustive. Thus, it is possible that VR encompasses more services than what was given as options in the questions to the respondents, although the category “Other” was provided as a choice. While this is not the main objective of this study, information relating to VR practices may provide insight to the variety of services provided under VR in different countries and, as such, may indicate the scope of VR and the professionals who provide it. Closely examining the category “Other” revealed some redundancy with VR services that were already specified. Finally, our study does not provide ways on “how” to measure the ICF categories. We feel that this is a critical next step if we are to operationalize the ICF categories in actual VR setting.

Conclusions

Our findings support the wide array of factors on functioning domains, from the experts’ perspective, that need to be considered in VR practice and research. This study has provided us with a list of ICF categories that were considered to be important in the VR process by expert respondents and which can help advance our understanding of the factors towards successful outcome. We encourage the VR community to further examine the list of variables provided here. Validation studies are needed in the near future to look at the utility of the categories in interprofessional communication (i.e. health care providers), service provision and reimbursement, and health information record.