LINKING TINNITUS HANDICAP INVENTORY - THI WITH THE INTERNATIONAL CLASSIFICATION OF FUNCTIONALITY, DISABILITY AND HEALTH – ICF: AN EXPLORATORY STUDY FOR THE APPLICATION OF THE LINKING RULES OF THE WORLD HEALTH ORGANIZATION

The International Classification of Functioning, Disability and Health – ICF is an important instrument to describe the functionality of people with tinnitus. The Tinnitus Handicap Inventory (THI), an instrument for the characterization and quantification of tinnitus, and due to its practicality and diverse application possibilities, has already been translated into several languages. This qualitative, descriptive, and exploratory study intends to present the application of the THI linkage rules with the ICF, and to propose the improvement of techniques for interpretation of results, preparation of reports and multidisciplinary therapeutic planning according to the perspective and domains established by the ICF for use clinical and occupational.


Introduction
Assessing the impact of tinnitus on quality of life is essential for effective treatment with the patient. The Tinnitus Handicap Inventory (THI), an instrument for the characterization and quantification of tinnitus, and due to its practicality and diverse application possibilities, has already been translated into several languages. The instrument is of great value for helping to assess and determine the degree of severity of tinnitus, as well as for characterizing and quantifying its impact on the patient's quality of life through questions that assess three dimensions: functional, emotional and catastrophic reactions to the tinnitus 1 .
The International Classification of Functioning, Disability and Health (ICF) is an important instrument to describe human functionality and disability. Its proposal prioritizes evaluating the impact of the symptom/pathology on the subject's functional life and not only his/her attention to the disease. From it, it is possible to describe the impact of tinnitus on a person's life, as well as its consequences on their functionality. The biopsychosocial model promoted by the ICF enables reflection on a multidimensional care approach, as well as the organization of health data as important information 2 .
The ICF represents an advance in the classification of disabilities, by allowing a variety of uses, including in the health area, with implications and direct use in social security policy and the formulation of public policies. In addition to adding the possibility of changing the health paradigm, it modifies the form of care, previously restricted to the causality of diseases, to consider its impact on people's quality of life 3,4 The first stage of the connection process was carried out according to the guidelines of the reference methodological guide10. The THI information was extracted and registered in a table with all the results of the linking process (Table 1).
In the second stage, after selecting the codes for each of the 25 items, codes were also selected to represent and classify the score of each subscale. This selection took place in accordance with the guidelines of Annexes 2 and 3 of the ICF. The extraction of the content and selection of codes for the subscales, as well as the linking of the results with the qualifiers, were also registered in table 2.
To verify the degree of agreement of the selected codes with the link proposed by James (2018) 11, an inter-examiner reliability analysis was performed using the Cohens Kappa coefficient as recommended by the linking rules. The Kappa correlation coefficient varies between 0 and 1, where 0 indicates no agreement and 1 indicates perfect agreement. This test is often indicated in the literature as the ideal test to assess the degree of agreement among RECIMA21 -REVISTA CIENTÍFICA MULTIDISCIPLINAR  ISSN 2675-6218   LINKING TINNITUS HANDICAP INVENTORY -THI WITH THE INTERNATIONAL CLASSIFICATION OF FUNCTIONALITY, DISABILITY AND HEALTH -ICF: AN EXPLORATORY STUDY FOR THE APPLICATION OF THE LINKING RULES OF THE WORLD HEALTH  3 professionals in identifying significant concepts and in linking these concepts with the ICF categories12.
Statistical analysis was performed with SPSS 22.0 software and obtained a highly significant result with a value of k = 0.91, with no missing data in the comparison between the two results. This result demonstrates that this proposal is as reliable as James' proposal11 with regard to linking the 25 items, as the author restricted himself to identifying the items individually, and it is still necessary to carry out the proposal for linking the subscales using the qualifiers for the effective classification of the impact with regard to the three domains assessed by the scale.
In the third stage, the selected codes were transported to the THI instrument, in order to identify each item linked to the ICF. Identification columns were created as an interface for identifying both the items and the results. Still as an interface for linking the results, a table was created to classify the subscale codes and the general score with their respective qualifiers, plus a specific space for the conclusion of the results (Table 3). More information about the linkage proposal will be provided below.

Linking Rules with the ICF
The use of the International Classification of Functionality in conjunction with clinical practice tools is highly recommended. Linking instruments with the ICF is a way of valuing the information already collected by the professional, mapping and complementing processed information and, at the same time, encouraging professionals to select the construction of more appropriate instruments for their clinical practice. However, sometimes, due to lack of mastery of the proposal, these tools are used from a completely different perspective from their proposal with the biopsychosocial model and remain anchored in a biomedical health model, despite including data on social participation and environmental conditions. Approaches only from this perspective result in decontextualized practices, strictly focused on interventions and monitoring only of signs and symptoms12.
To correctly guide this procedure, a methodological guide was published in order to avoid diversified practices. For this reason, in 2002 the first proposals for the connection rules were presented, which were updated in 2005 and refined in 2016 by Cieza10. Among the main guidelines, the need to identify the main proposal of the information to be linked is highlighted before identifying the significant concept. The rules propose much more than just identifying a significant concept, but also separating the content of the item as the main concept and additional concepts, according to the proposal and perspective of the information to be connected with the ICF. The recommendation to link the content of the instruments with the ICF, so that the ICF is the tool capable of translating the information contained in the instruments and not the other way around. Associating validated instruments with the ICF categories is not an easy task, however, international efforts have been applied to assist researchers in this process. The use of the ICF equips professionals in clinical practice, and its use can help in the selection and construction of instruments that use more accurate content.
Linking ICF content with clinical practice instruments can contribute to the operationalization of an expanded concept of health, allowing the same degree of importance to be attributed to psychological components and contextual factors, considering privileging their interaction as a health product and producer and its related states.
The publication of the latest linkage rules intends to maximize the transparency and reliability of the process between the instrument's content and the components, and for this, improvements were proposed that highlight the need for a thorough understanding of the instrument, the model and taxonomic foundations of the ICF . Another guideline currently advocated is regarding the process of documenting the perspective contained in data collection and the category of responses (frequency, intensity, duration...) to relate the concepts involved in the responses as well.
Cieza et al. 201610, guides the realization of a detailed description in the elaboration of the linking document, since it is necessary for the process of linking the information to describe the perspective analyzed at the time of linking the categories. The recommendation of a list of ICF categories addressed is reinforced (Annex 1) and the record of the perspectives from which the information was collected, and specifically in the context of instruments or assessments, also in the approaches taken to categorize response options. The author emphasizes that the information will provide specialists, professionals and researchers with a more comprehensive and informed view to decide whether health information is comparable or not, if it is comparable with the perspective and categorization, in principle it is considered adequate for aggregation and quantitative comparison . In this way, the character of standardization and unification of the language by the ICF can be strengthened and fulfill its objective.
Although the linking process can be applied to any category of health information, it is not always feasible to link certain information to a specific ICF category. This occurs with information that we intend to link, but they are and in extreme opposites (beyond the scope of the ICF or too specific to be covered by the alphanumeric code). The practice suggested in publications that guide the bonding process in 2002 and 200510 refers to the use of the abbreviation nc (not covered) as a possibility at the time of decision in challenging cases for the ligament. In this study it was not necessary to use this resource, all 25 items and subscales were coded. The non-coding of the result for the general score is not due to the absence of an adequate code, but to the understanding of functionality as a complex construct and, consequently, composed of the sum of the codes of the subscales, the domains of the THI, as we will discuss later. All three versions were analyzed together with the original version of the instrument8. Then, a synthesis of the content was extracted, its connection with the components of the ICF for each item of the scale, as well as for the score of each subscale and their respective qualifiers (Chart 1 and 2). The selection of components for each item was identified in a column in the Ferreria (2005) version. The attached version was selected as the first publication in the language, as well as being mentioned by the original author of the scale in one of the international reference documents that also make up this work (Appendix 2),

Tinnitus Hadicap Inventory -THI
Regarding psychometric properties, the THI showed high correlations between items involving boredom, sleep disturbance, depression and concentration. This is one of the findings that suggests adequate construct validity for those disabling reactions resulting from tinnitus. The symptoms distributed in the instrument's subscales represented 45% to 52% of the variance in 5 the total score of the scale, which suggests the respective classification for interpretation of the instrument's general result.

RECIMA21 -REVISTA CIENTÍFICA MULTIDISCIPLINAR ISSN 2675-6218 LINKING TINNITUS HANDICAP INVENTORY -THI WITH THE INTERNATIONAL CLASSIFICATION OF FUNCTIONALITY, DISABILITY AND HEALTH -ICF: AN EXPLORATORY STUDY FOR THE APPLICATION OF THE LINKING RULES OF THE WORLD HEALTH ORGANIZATION
The instrument consists of 25 items distributed into three subscales. The Functional Scale with 11 items and scores between 0 -44; Emotional Scale with 9 items and a score between 0 -36 and the Catastrophic Scale with 5 items and a score between 0 -20. According to the guidelines for applying the rules of connection with the ICF, the main concepts and additional concepts with their respective interpretation and alphanumeric selection for each item. Given that the gathering of items from a subscale represents a domain of functionality through a specific score (F, E and C), it is important to remember that each of these escores provide qualitative information about functionality while the overall scale score results , in fact, from the sum of the three domains (three subscales) and generates a score with a quantitative representation of functionality. Often the results of subscales are completely underestimated due to the overall score, which would be a complex construct. For this reason, in this linking work, we propose to link both the individual items and the domains of each subscale. In this way, there will be the possibility for the professional to individually classify each aspect or domain (qualitative analysis) and its impact on functionality (general score), regardless of a preserved functionality or some degree of disability (activity limitation or participation restriction) revealed through the final score > 50%, as suggested by the distribution of variance described by the authors in the original article.
It was found that all 25 items were in the first part of the ICF referring to the concept of Functionality (positive aspect of the classification) or Disability (negative aspect of the classification). Regarding the linking of the score of each subscale, the assigned codes were also identified in the first part of the ICF and linked to a second level code, as suggested in the case of application in health research and evaluation, in Annex 2 of the ICF4. This classification proposal will be described in detail later.
The first part is composed of two components, the first Functions (FC) and Structures of the Body (EC) and the second Activities and Participation (AP) of the individual. Both are divided into four constructs that make up the assessment of functionality and disability: changes in body functions (b); changes in body structures(s); capability (executing tasks in a standard environment) (d); performance in execution (executing tasks in a habitual environment) (d). Each component has, and its respective constructs are classified through the use of qualifiers that complement a unique alphanumeric code identified by the initials of words such as b (body), s (structure), d (domain) and the letter e (from English environment).

Interpretation and Classification of THI Results
As mentioned earlier, the Tinnitus Handicap Inventory was created by Newman et al. in 1996 and became one of the most used self-report measures to measure the functional impact caused by tinnitus in people's lives. In addition to also measuring the effects of treatments, the dynamic character of the THI scores enables verification of the dynamics of life related to functionality or human disability resulting from tinnitus expressed through a score obtained between 0 and 100. As it is a psychometrically robust tool, there is a consensus in the literature that the general result obtained by the instrument reveals the level of commitment proportionally to the scores presented.
Although the THI was originally published in the United States in 1996, the authors decided to make changes to the instrument's total score in 1998 and established severity categories based on calculated quartiles with the classification 0-16 without disability, 18-26 mild disability, 38-56 6 moderate disability and 58-100 severe disability. In 2001, a group of British researchers proposed the reorganization of the categories based on five groups, with 0-16% considered negligible, 18-36% mild, 38-56% moderate, 58-76% severe and 78-100% catastrophic13. Although this classification was not originally part of the instrument, even with the maintenance of the identification of subscales, since then its classification has been exclusively linked to the quantitative interpretation of the instrument's total score over these years. This practice is reflected in the absence of identification of subscales in some translations of the instrument into other languages. In clinical practice, this habit can even generate some disregard for qualitative aspects through the importance or relationship of subscales E, F and C in understanding their three-dimensionality related to the impact of tinnitus, as proposed by the instrument itself.
For Tunkel et al. (2014) classification schemes offer an approach for professionals to measure changes in treatment over time, with and without intervention. This method provides the researcher with an opportunity to assess group differences using non-parametric statistics14.
The authors also point out that some questionnaires have been used to document problems resulting from tinnitus, as well as to measure changes in tinnitus with treatment and differ, among other things, in measurement scales for the main functions and secondary activities affected by tinnitus. Exactly from this perspective, we resume that the ICF has as its main objective the standardization and unification of the language between professionals and their areas of expertise, in order to enable the obtaining of evidence and statistical studies with international standards. For this purpose, as the name itself describes the International Classification of Functioning, it can be considered a tool that makes it possible to obtain a common language and has been recommended in official documents of international reference groups in research in the tinnitus area.
The proposal for Classification of the Functioning of people with Tinnitus is in line with the THI proposal, mainly with regard to the analysis of content in a quantitative, but also qualitative way, and enables the mapping of the aspects of the greatest impact arising from the perception of Tinnitus. Thus, by offering both quantitative and qualitative understanding of the repercussions of tinnitus, verification and understanding of the multidimensionality that makes up human functionality. The ICF also allows for more effective and efficient therapeutic planning when linked to assessment instruments such as the THI. As an official and international instrument, the ICF adopts its own classification that is premised on the universalization and equivalence of results through research and adopts as classifiers the broad classes of percentages for cases that can be used both for calibrated instruments and for other standards of measurement of impairments, capacity limitations or performance problems related to functionality. Their percentages are calibrated for different domains with reference to population patterns as percentiles.
In order to carry out this exploratory study, as well as to enable the implementation of the Linkage proposal between the results of the assessment instrument (THI) and the classification (CIF), for several reasons that will be presented below, it is suggested to use the classification of the CIF and its qualifiers for the interpretation and results obtained with the THI. The theoretical foundation and information on disability values, as well as the psychometric properties studied regardless of their link with the classifications currently used for interpretation of the general score, and use of the integral and qualitative proposal of the instrument with information offered by the respective subscales, added to the ample evidence for classification of functionality, follow the WHO recommendations for its use and, consequently, the proposal of linking the instruments.

Alphanumeric Codes and Qualifiers
The classification of the ICF components is carried out by the formation of an alphanumeric code that includes the presence of the qualifiers. It is important to highlight that these are only complete with the presence of at least one qualifier after the one separating point (xxx.1). This means that without the presence of a qualifier, the code completely loses its meaning.
According to WHO4, the function of the qualifier is to indicate the magnitude of the extent and severity of the problem, or the level of health. All ICF components are quantified using the same generic scale represented by:

Categorization of Subscales and use of Activity and Participation list for distinct Domain Groups without overlapping
Regarding the categorization of the functional subscale, the d720 category was initially selected, however, it was found that Annex 3 of the ICF presents a complete list of Activity and Participation domains. It is considered that this list completely covers the range of functionality that can be encoded, both at the individual level and at the social level. Given this possibility, it was decided to use the list of Activity and Participation domains for distinct groups without overlapping. There was grouping of categories for the use of only one code for a group of categories related to Activities (a) or Participation (p). The Functional Subscale was categorized with code p7 (Interpersonal interactions), also having as a suggestion the use of p8 (Main Areas of Life), mainly for occupational assessment situations.
Regarding the description of the results, the classification words must be chosen according to the classification domain. In the specific case of the THI, we will have a qualitative result (identified of the most affected aspects from the subscales) and a quantitative result (general score).
The result of the subscales for the Emotional and Catastrophic scales correspond to body functions (b152 and b160), both reveal Functional Integrity (or without any difficulty) in case of qualifier (.0), or Deficiency for b152 (emotional functions) or b160 (thought functions), followed by the degree corresponding to the value obtained in these subscales. For the Functional Subscale (F), the result corresponds to a Capacity Limitation or Performance Problem for the performance in Activities and Participation (codes p7 and p8), followed also by the degree of perceived impact. As for the classification of the overall score, it is suggested, in case of values 12 the THI with the ICF biopsychosocial model, it is possible to map the most compromised aspects and suggest treatments that are more adequate to the patients' needs.

RECIMA21 -REVISTA CIENTÍFICA MULTIDISCIPLINAR ISSN 2675-6218 LINKING TINNITUS HANDICAP INVENTORY -THI WITH THE INTERNATIONAL CLASSIFICATION OF FUNCTIONALITY, DISABILITY AND HEALTH -ICF: AN EXPLORATORY STUDY FOR THE APPLICATION OF THE LINKING RULES OF THE WORLD HEALTH ORGANIZATION
The link between the THI and the ICF and the use of its qualifiers provide an opportunity for a qualitative and quantitative look at the impact of tinnitus on people's lives, however, without restricting tinnitus to one-dimensionality, as occurs in the use of an exclusively quantitative classification of results. The linking procedure between the instruments made it possible to relate the ICF codes and qualifiers with each item of the scale independently, as well as to classify the dimensions represented by the correspondence with each of the subscales.
The reliability analysis revealed that this proposal proves to be as reliable as James' proposal11, but it presented some innovations such as the link and classification of subscales, with a new interpretation for the overall THI score. The author also emphasizes that the severity classification currently proposed is not sufficient to contemplate the activity limitation and participation restriction in tinnitus, and that complements and states that the composite scores of the assessment scales can hide the actual affected areas of tinnitus functioning. This statement corroborates the proposed interpretation of the results.
Furthermore, despite a few items to assess Activity and Participation, it is clear that the correspondence of the general result in measuring the impact of tinnitus on human functionality can be classified in its quantitative aspect as disabling when there is a superior result to 48%, as suggested by the authors in the original article.
It is clear from this that the ICF structure precisely identifies the affected areas in the lives of tinnitus patients, based on which an intervention model can be prepared for application in groups and individuals. Such models act as a facilitator for symptomatic intervention, as the perceived problem due to tinnitus is a subjective factor that varies between individuals, depending on their coping and defense mechanisms used, personality traits and personal attitudes towards tinnitus.
Finally, the link between the instruments and the correct use of the classification table with their respective qualifiers, enable health professionals to prepare Occupational Reports and Reports that reflect the impact of tinnitus and its repercussions on various aspects of life, including work of people with disabilities resulting from the presence of tinnitus. It is suggested to use the THI in partnership with WHODAS 2016 when values are greater than 48% and that other studies based on the same proposal be carried out to verify, refine and validate this proposal as an alternative for interpreting the THI result in agreement with the ICF. It is expected with this to help professionals and multidisciplinary teams in managing and directing the most appropriate treatments for cases of people with tinnitus.   LINKING TINNITUS HANDICAP INVENTORY -THI WITH THE INTERNATIONAL CLASSIFICATION OF FUNCTIONALITY, DISABILITY AND HEALTH -ICF:  AN EXPLORATORY STUDY FOR THE APPLICATION OF THE LINKING RULES OF THE WORLD HEALTH ORGANIZATION  Scheila Farias de Paiva