Integrative biopsychosocial model of functioning, disability, and health.
Tschiesner U, Becker S, Cieza A. Health Professional Perspective on Disability in Head and Neck Cancer. Arch Otolaryngol Head Neck Surg. 2010;136(6):576-583. doi:10.1001/archoto.2010.78
To evaluate problems after head and neck cancer (HNC) from the multidisciplinary team perspective; to classify the results using the International Classification of Functioning, Disability, and Health (ICF); and to compare the results with a patient perspective.
There were 103 participants from 27 countries: 50 physicians (otolaryngologists, maxillofacial specialists, and radiation and medical oncologists) and 53 nonphysicians (dentists, psychologists, physiotherapists, speech swallowing therapists, nurses, and social workers).
Health professionals involved in the treatment of HNC were asked about relevant problems. The survey was Internet based and included 5 questions, 1 for each of the ICF components: Body Functions, Body Structures, Activities and Participation, and contextual Environmental and Personal factors. Answers were translated into ICF categories by 2 independent researchers, and frequencies were calculated. The results were compared with the outcomes of patient interviews based on similar questions.
A total of 3643 different answers translated into the ICF using 160 different second-level ICF categories. Less than 1% of answers were not covered by the ICF. There was high consistency in the ratings of food ingestion, pain, and the relevance of the immediate family. In general, health professionals tended to emphasize aspects of anatomical defects and body image, whereas areas of speech and exercise tolerance functions were more often named by patients.
The ICF seems to be a comprehensive tool for classifying problems after HNC from the multidisciplinary health professional perspective. There are important differences between the health professional and patient perspectives. We should be aware of this during cancer follow-up sessions and in the creation of rehabilitation plans.
During the past few years, there has been increasing interest in and awareness of the behavioral and functional impact of oncologic treatment on the patient, and health-related quality of life has become a secondary end point in the assessment of outcome.1- 8 However, it is often difficult to compare results of the studies because they refer to different standards and study aims and often are not comprehensive.9 Some authors10,11 suggest a uniform language and methods to improve utility information in head and neck oncology.
In May 2001, the World Health Organization (WHO) adopted the International Classification of Functioning, Disability, and Health (ICF)12 as a common framework for all health professions12,13 and as a unified and standard language for the description of health and health-related states in rehabilitation. In the clinical context, the ICF is intended for use in needs assessment, matching interventions to specific health states, rehabilitation, and outcome evaluation.14
The ICF model of functioning does not see a patient's functioning in life as a mere consequence of disease but describes functioning in the context of a comprehensive biopsychosocial network with its components body functions (b), body structures (s), and activities and participation (d) as well as individual contextual personal and environmental factors (e). Each component consists of several chapters, and within each chapter are categories, which are the units of the classification (Figure).12
The ICF helps to comprehensively address and structure patient problems in a multiprofessional teamwork approach, which many times is required when dealing with head and neck cancer (HNC), where fighting the tumor and managing the sequelae after therapy go together. It often requires coordinated longitudinal care involving dentists, nurses, physicians, physical therapists, psychologists, speech and language therapists, social workers, and others. Because the ICF classification provides shared terminology for all health professions, it is a promising tool for facilitating the multidisciplinary working approach in HNC. Shared terminology might also facilitate international multicenter studies and ease the comparison of study outcomes, even if different outcome measures are applied.15- 17 However, it has not been determined yet to what extent the ICF contains all the aspects of functioning and disability that are considered relevant by health professionals involved in the treatment of HNC.
The health professional perspective has an important effect on treatment decisions because health professionals inform and advise the patient. However, although the patient perspective is generally covered by a variety of well-validated patient-based questionnaires18- 20 and the academic researcher perspective is covered by a variety of accurate literature reviews, very few studies assess in greater detail the perspective of academic and nonacademic health professionals involved in the treatment process.21- 25 It is crucial to understand the health professional perspective and to compare it with, for example, the patient perspective. This understanding might help us to further challenge and improve treatment regimens and to work against professional bias.
Therefore, the aim of this study was to systematically study the health professional perspective on what factors are relevant to patient functioning in life and to check whether these factors are included in the ICF. To study the health professional perspective as comprehensively as possible, information needs to be collected from (1) the different health professions involved in the treatment and (2) different WHO world regions to include different ethical, social, and political backgrounds. The ICF classification was used to translate all health professional answers into the same neutral “language” and, thereby, facilitate comparison of the given answers.
The specific aims were (1) to link the concepts contained in the health professional answers with detailed ICF categories and to determine the frequency for each ICF category, (2) to analyze which concepts are not covered by the ICF, and (3) to compare health professional answers in this survey with patient answers in an earlier study26 based on a similar set of open-ended questions.
In an Internet-based survey, health professionals from various cultural backgrounds and different health professions involved in the treatment of HNC were asked 5 open-ended questions about what factors were relevant to their patients. The answers were given in plain English and afterward were translated into ICF categories by 2 independent researchers according to a predefined “linking process.” Further analyses were performed on the basis of ICF categories. Answers that could not be linked to detailed ICF categories were grouped by content.
Health professionals from all 6 WHO world regions were included: Africa, North and South Americas, Eastern Mediterranean region, European region, Southeast Asian region, and Western Pacific region. Health professions included in the study were dentists, nurses, physicians, physiotherapists, psychologists, speech and language therapists, and social workers. Physicians included oto(rhino)laryngologists, maxillofacial and oral surgeons, and radiation and medical oncologists. Other inclusion criteria for participants besides professional background were professional experience with patients with HNC for at least 5 years, a fluent command of the English language, age older than 18 years, and informed consent to contribute to the survey.
In the preparatory phase of the study, professional associations of health professions engaged in the treatment of HNC worldwide were identified by means of an Internet search and were contacted. In addition, another Internet search and personal recommendations were used to identify single health professionals experienced in the treatment of patients with HNC.
A list of nominated health professionals was created. The list covered the selected health professions (aim: 50% physicians and 50% nonphysicians) and all WHO regions (aim: almost equal distribution). These health professionals were contacted via e-mail, were informed about the study, and were asked to participate. Each health professional who did not respond within 2 weeks received a reminder.
All health professionals who met the inclusion criteria and agreed to participate in the survey received individualized access (user name and password) to enter the Internet-based survey platform. There they found instructions on how to proceed and 2 questionnaires. The first questionnaire asked for personal information to verify the inclusion criteria. Personal information included age, sex, health profession and specialties, country in which the person was working, and years of professional expertise. The second questionnaire consisted of 5 open-ended questions asking the health professional to name relevant changes due to the cancer and its treatment. The questions covered the ICF chapters in body functions, body structures, and activities and participation. Additional questions related to contextual environmental and personal factors that affect life (Table 1). For each open-ended question, a health professional could write down as many items as he or she considered relevant. Participants could not see each others' answers. Open-ended questions were applied to avoid any bias included in predetermined closed-ended questions. All correspondence was in English.
After receiving answers in plain English from all the participating health professionals, the answers were translated (linked) with the ICF language based on established linking rules in a systematic and standardized way.27,28 According to these linking rules, each concept of functioning contained in the answers was linked to the ICF category representing this concept most precisely. An answer could be linked to 1 or more ICF categories, depending on the number of themes contained in the answer. To overcome the possible bias of any health profession, the linking was performed by 2 different health professionals: a physician (S.B.) and a psychologist. The linking approach has been proved to be useful for translating any kind of health information to the ICF,29- 31 including HNC.17,26,32,33
Not all answers can be linked to detailed ICF categories, but they still have to be documented:
Personal factors belong to the biopsychosocial concept of the ICF. Personal factors refer to age, sex, ethical background, living status, socioeconomic status, smoking habits, alcohol consumption, individual coping strategies, self-esteem, and individual knowledge of disease and treatment options. However, they are not yet specified into detailed ICF categories. Therefore, all personal factors are labeled “pf” without further specification possible.
Concepts that deal with the underlying health condition (cancer diagnosis and its treatment) cannot be linked to the ICF. They can be coded in the ICF instead. All such concepts are labeled as “health condition” (hc).
Concepts that are too general and not precise enough to be grouped into an ICF category are classified as “not definable” (nd).
Concepts that deal with functioning but that still are not represented by the ICF are labeled as “not covered” (nc). Such concepts may represent concepts that are outside the scope of the ICF.
A peer review was performed for the linking procedure. Random samples of 50% of the answers were linked by a second health professional. The degree of agreement between the 2 health professionals regarding the linked ICF categories was calculated using the κ statistic, with 95% bootstrap confidence intervals.34,35 The values of the κ coefficient generally range from 0 to 1, where 1 indicates perfect agreement and 0 indicates no additional agreement beyond that expected by chance alone.
κ Statistics were calculated per component at the second ICF level to indicate the degree of agreement between the 2 independent linkage versions of the 2 health professionals. If a concept of a parameter was linked to a third- or fourth-level ICF category, the overlying second-level category was considered. The ICF is organized in a hierarchical scheme so that the more specific lower-level categories share the attributes of the less specific higher-level categories.27
A category was counted only once for each health professional, even if several answers were given that were all translated into the same ICF category. Using this approach, the results reflect the number of health professionals who consider an ICF category relevant for patients with HNC and are not biased by, for example, few health professionals who name the same concept repeatedly. All answers were kept anonymous.
In an earlier study,26 patient interviews were held using the same set of 5 open-ended questions applied in the present survey. The patients had carcinomas of the oral region, the hypopharynx, and the larynx and an average survival of 5.4 years. Eighty-three percent of patients underwent a surgical procedure of the primary tumor, 53% underwent neck dissections, and 26% were treated with postoperative radiochemotherapy. Conversely, 17% of patients were treated nonsurgically with primary radiochemotherapy. The interviews were fully recorded, typed, and analyzed. Answers were translated into ICF categories according to the same procedure and rules as applied in this study. They were reported also at the second level of the ICF classification. Data analysis was performed using a statistical software package (SAS for Windows version 9.1; SAS Institute Inc, Cary, North Carolina).
Four hundred twelve health professionals were identified directly by means of an Internet search or on the basis of personal recommendations of other participants (snowball effect); 132 health professionals (32%) answered. The reasons for not answering are unknown. They include default e-mail addresses, spam filters, and lack of interest in the survey. In general, physicians and speech and language therapists had a higher rate of response (>40%) than did psychologists and social workers (<10%).
One hundred eleven of the 132 health professionals (84%) fulfilled the inclusion criteria, agreed to participate, and logged in to the Internet platform. Owing to incomplete answering of the questionnaire, the answers of only 103 health professionals could be used for further processing. Reported reasons for incomplete answering were “being too busy” (n = 3) and “did not understand the questionnaire” (n = 1).
Participating health professionals had a median age of 45 years (age range, 24-70 years) and have been working in the field of HNC for a median of 16 years (range, 5-37 years). The male to female ratio was 54:49.
The 103 health professionals belonged to 7 different health professions: 50 were physicians and 53 were nonphysicians. The nonphysicians included 17 speech and language therapists, 10 physiotherapists, 7 nurses, 7 dentists, 6 psychologists, and 6 social workers. In the physician group, there were 18 oto(rhino)laryngologists, 7 radiation oncologists, 6 medical oncologists, and 10 oral and maxillofacial surgeons. Nine physicians did not specify their work field further.
In terms of the WHO world regions in which they practice, there were 48 participants from the European region (Belgium, Denmark, France, Germany, Greece, Israel, the Netherlands, Portugal, Spain, Sweden, Switzerland, and the United Kingdom), 21 from the Americas (Argentina, Canada, and the United States), 18 from the Western Pacific region (Australia, China, Korea, and Malaysia), 6 from Southeast Asia (India), 6 from Africa (Kenya, Namibia, Nigeria, and South Africa), and 4 from the Eastern Mediterranean region (Iran, Oman, and United Arab Emirates). Each participant gave a mean of 39 (range, 2-145) different open-ended answers.
All text answers included 4292 aspects of functioning given by the health professionals, and 3643 answers (85%) could be linked to 160 different ICF categories. The ICF classification allows for differentiation of categories according to their distribution into components of the biopsychosocial model. Of the 160 identified ICF categories, the survey shows a proportional distribution of 47 body functions (29%), 28 body structures (18%), 49 activities and participation (31%), and 36 contextual environmental factors (23%).
Six hundred forty-nine answers (15%) could not be linked to detailed ICF categories for 1 of the following reasons:
A fifth component of the ICF is the contextual personal factor. However, personal factors are not yet classified into detailed ICF categories by the WHO. Therefore, concepts that refer to this area cannot be translated into detailed ICF categories yet and are labeled “personal factor” instead. Altogether, 386 answers (9%) from 91 participants (88%) referred to the personal factor component. In this group, the most prominent aspects were related to (1) smoking and alcohol use, (2) individual coping strategies, and (3) socioeconomic status.
Another 107 answers (3%) from 54 participants (52%) covered aspects of the underlying health condition and, consequently, were labeled “health condition.”
All answers that were too general to be linked to detailed ICF categories, such as “physical limitations,” “decline in quality of life,” and “symptom management,” were labeled “not definable.” One hundred twenty-one answers (3%) from 67 participants (65%) were labeled “not definable.”
However, there were answers that referred to detailed aspects of functioning and disability that are not covered by the ICF. This group of answers needs special attention. Altogether, there were 35 different answers (<1%) given by 22 participants (21%) that referred to the following aspects: (1) mucus production and lung secretions, (2) alterations in dentition, (3) sleeping position, (4) education of family members and colleagues, and (5) general public awareness. All of these answers were labeled “not covered.”
Interlinker concordance was 0.723 (95% confidence interval, 0.692-0.754).
For each ICF category, it was calculated how many health professionals had named it. Altogether, there were 160 different second-level ICF categories. Of these, 65 categories were named by at least 10% of the participating health professionals (Table 2). The most prominent ICF components were body functions and activities and participation; within these components, ingestion function (b510) (83%) and eating (d550) (58%) were most relevant to the health professionals.
Summarizing the results of both studies, 171 categories were identified altogether (Table 3). The health professional survey came up with 160 categories and the patient interviews with 104 categories. Ninety-three of 171 categories (54%) were identified in both studies, and 67 categories (39%) were exclusively named in the health professional survey. Another 11 categories (6%) were mentioned exclusively by the patients (Table 4).
Despite the fact that health professionals came up with a greater variety of aspects than did patients, there were many similarities between the 2 groups. There was consistency, especially in the fields of food consumption (b510, d550, and d560), sensation of pain (b280), importance of the immediate family (e310), and the relevance of products for personal use in daily living, covering tracheotomy cannula and percutaneous endoscopic gastrostomy tubes (e115).
However, some differences need special attention. Health professionals considered some aspects more relevant than did patients. Body image, as summarized under b180, was named by 46% of health professionals and just 6% of patients. The extended family (e315) and economic self-sufficiency (d870) were named by clearly more health professionals than patients. Anatomical changes and defects regarding the mouth (s320), pharynx (s330), muscles and ligaments of the head and neck area (s710), the skin (s810), and the cranial nerves (s110) were also named by substantially more health professionals than patients. Also, categories related to adverse effects of radiochemotherapy, such as hearing functions (b230) and repair functions of the skin (b820), were named more often by health professionals than by patients.
In contrast, other aspects were named by patients more frequently. In general, many aspects of speech (b310, d330, d350, d360, d345, and d355) were identified more often by patients than by health professionals. Exercise tolerance function (b455) was identified by 44% of patients but just 6% of health professionals. Also, emotional factors; substances for consumption, such as food and drugs (e110); and the climate (e225) were more often named by the patients (28% vs 2%).
This survey collected the perspective of health professionals experienced in the treatment of HNC. It deliberately integrated information from different health professions and from various cultural backgrounds as determined by different WHO world regions and countries. Doing this, the ICF was used to translate multiple answers into a neutral language. The survey could show that the ICF contains most aspects considered relevant by health professionals. On a more general level, the ICF has proved again to be a practical tool for translation of problems encountered after HNC.17,26,36 It allows for the comparison of study results collected in different regions of the world even if different study designs and outcome measures have been applied.14,15,30,37 This means that there is hope that the ICF might assist us in reducing the data silo effect,9 instead getting full benefit out of the results of different clinical studies worldwide.
Very few answers (<1%) referred to aspects of functioning and disability that are not covered by the ICF. This underlines the relevance of the ICF for HNC. The identified aspects of mucus production, altered dentition, sleeping position, and education of families and peers are not covered by the ICF and need to be collected carefully and reported for possible inclusion into a future update of the ICF.
However, one area of the ICF model is not detailed enough yet. According to the WHO biopsychosocial model of functioning, there is a fifth component, the so-called personal factors. During this international multiprofessional survey, most health professionals (88%) mentioned aspects that refer to personal factors. Further detailing of the personal factor component into detailed ICF categories seems most appropriate to further increase the relevance of the ICF classification in different health conditions, eg, HNC.
The health professional view was determined with the inclusion of approximately 50% physicians (with surgical and oncologic approaches) and 50% other health professionals, including speech and swallowing therapists, physical therapists, psychologists, dentists, nurses, and social workers.
This survey showed similar ratings for categories such as eating and pain sensation. However, there seem to be other aspects of functioning and disability, such as body image and anatomical changes, that might be overrated by health professionals compared with the patient perspective, whereas other aspects, including speech, exercise tolerance functions, and emotional functions, are relatively underrepresented. Such differences might be affected by the distribution of health professions in this survey. It includes more ears, nose, and throat and maxillofacial surgeons (n = 28) and relatively fewer speech and language therapists (n = 17), physiotherapists (n = 10), and psychologists (n = 6). According to this study, early integration of speech and language therapists, physiotherapists, and psychologists in the multiprofessional treatment team might improve addressing patient complaints in rehabilitations plans and improve patient quality of life in the long run.
Categories that relate to adverse effects of aggressive radiochemotherapy, such as hearing functions and repair functions of the skin, seem to be more often identified by health professionals than by patients. This, however, can be explained with patient selection in the patient interviews, which included less than 20% of patients after primary radiochemotherapy.
This study was subject to some limitations. It was designed to cover perspectives from different professional and cultural backgrounds. Although the aim of approximately 50% physicians and 50% nonphysicians was accomplished, an equal distribution in terms of WHO regions could not be reached. Although health professionals in various countries had initially been contacted, within the participants there is predominance toward Europe and North America. In relation to this, especially the Eastern Mediterranean region (reflecting Iran, Oman, and United Arab Emirates) is underrepresented. The reasons for this are unknown; however, they might be connected to the inclusion criteria of, for example, “fluent command of the English language,” even in countries where English is not the first language and possibly not taught to a great extent in school. Such inclusion criteria might keep experienced health professionals from participation. Other reasons might lie in free access to the Internet to be contacted or in the personal interest to cooperate with a study center from a cultural background located in Germany. All this confines the conclusions drawn from this study to the American and European regions and restricts the informational value toward conclusions for the other WHO regions. In addition, a high percentage of health professionals had been contacted via e-mail but never answered. Because the reasons for not answering are unknown, any bias resulting from this cannot be excluded.
This survey was an attempt to collect the views of a multiprofessional team on what are the relevant areas encountered after HNC and its treatment. Analyses were based on the ICF. The ICF seems to be comprehensive enough to address the perspective of different health professions. A comparison with the patient perspective was created by linking the results to the ICF categories in both studies. This might give us some first insights regarding differences in the health professional and patient perspectives and how the health professional perspective might affect advice given to the patient and treatment decisions. However, more detailed research is needed.
Correspondence: Uta Tschiesner, MD, Department of Otorhinolaryngology, Head and Neck Surgery, Ludwig Maximilians University, Munich, Marchioninistr 15, Munich, Germany (email@example.com).
Submitted for Publication: July 2, 2009; final revision received November 10, 2009; accepted January 12, 2010.
Author Contributions: All authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Tschiesner and Cieza. Acquisition of data: Becker. Analysis and interpretation of data: Tschiesner and Becker. Drafting of the manuscript: Tschiesner. Critical revision of the manuscript for important intellectual content: Tschiesner, Becker, and Cieza. Statistical analysis: Tschiesner. Obtained funding: Tschiesner and Cieza. Administrative, technical, and material support: Tschiesner and Becker. Study supervision: Tschiesner and Cieza.
Financial Disclosure: None reported.
Funding/Support: This study was funded by the Deutsche Krebshilfe eV.
Additional Contributions: This study was a cooperative effort with the WHO. Edda Amman contributed to the linking procedure. We thank all respondents for their participation in this study.