Análise empírica do modelo teórico da Classificação Internacional de Funcionalidade, Incapacidade e Saúde (CIF)

Abstract

The International Classification of Functioning, Disability and Health (ICF) presents a classification system and a theoretical method based on the junction of medical and social models and uses a biopsychosocial perspective to integrate different health dimensions. In the ICF model, the functioning and disability processes result from the interaction between a Health Condition (HC) and Contextual Factors (CF) and is represented by three components: Body Function and Structure (BFS), Activity (A) andParticipation (P). The ICF model and its classification system have been used in the analysis of the functioning content approached by distinct assessment tools; in the development of elementary code groupings (core sets); in the analysis of the scientific evidence available on a certain theme; as a guiding model for clinical reasoning and professional performance; as well as to guide the development of curricular structures,health politics and guidelines for the health insurance and social welfare systems. Due to the extent and importance of this model, mainly as guidance for practice and research in rehabilitation, the measurement and analysis of its concepts and relations from empiric data are steps that require attention and should be developed. There are studiesavailable which have analyzed the inter-relations between the components of this model in a theoretical perspective, but few studies empirically explore this relation. Hence, its necessary to discuss the applicability of ICF in clinical practice and to consider which assessments are appropriate and inform about the concepts that support the model. Furthermore, the discussion about the influence of the contextual factors on the functioning process is latent in the researches that involve analysis of the model. Thus, the aims of this study were to: (1) select assessment tools to measure the different ICF components; (2) analyze the relations between its components; (3) test the empiricalvalidity of ICFs biopsychosocial model; and (4) analyze the influence of the contextual factors on the participation component. Two hundred and twenty six patients, with ages between 18 and 59 years, with different health conditions and who were in rehabilitation were assessed. In order to assess the ICF components, the following tools were used: WHODAS II, MIF, Participation Scale, CHIEF, a protocol with tests and various measures to measure BSF and a socio-demographic, occupational and life habit questionnaire, to characterize PF. For the analysis of the measures, tests and assessment tools, the factorial analysis by main components technique was used. The empirical testing of the ICF model and of the relations between its components was performed by structural equations modeling; while the regression analyses with hierarchical entry of data was applied to verify the association of the contextual and participation factors. This study reduced a group of information on PF to four independent factors, characterized by variables related to economic and study conditions; work situations, having children and living with a partner; smoking habits and alcohol intake; and sex and physical activity. These four factors showed a total explained variance of 59.1%. As for the BSF, there was a grouping of the information in three factors related to upper limb strength and pain intensity; lower limb strength; upper limb flexibility and body mass; and lower limb flexibility and sub-maximum effort. The total explained variance of the three factors of BSF was 61.3%. The assessments used to measure A, P and EF also presented a grouping pattern consistent with the literature, with explained variancesthat ranged from 60.5% to 68.0%. The model testing indicated good adjustments (Chisquare =295.6; gl = 63; GFI = 0.854; AGFI = 0.775; RMSEA = 0.028 [IC90% = 0.014- 0.043]) and showed a significant relation between BSF and A (standard coefficient = 0.32; p < 0.0001); A and P (standard coefficient = - 0.70; p < 0.0001); and CF and thefunctioning components (standard coefficient = 0.37 [p < 0.0001] and 0.34 [p = 0.00]). The direct effects between BSF and P (standard coefficient = - 0.10; p = 0.111) and SC and the functioning components (standard coefficient = - 0.12; p = 0.128) were not confirmed. Still in relation to the results, it was identified that more scholar years,having a job and drinking alcohol are conditions that increase the patients social participation. On the other hand, factors such as natural environment, transport, access to health services and social capital were perceived as the most important barriers to participation (Raj = 0.42 and p < 0.0001). These findings confirm the fact that the selected tests and assessment tools are alternatives anchored in the ICF model and couldbe used in rehabilitation centers. It can be inferred that changes that occur exclusively in BSF dont have a direct effect on P and the direct effect of SC on the functioning components was also not confirmed, which indicates that the other factors, such as those related to the physical and social environment and to the personal characteristics interactfor the production of the disability processes. Thus, this study points to a functioning model that comprehends a contextualized perspective and must also include variations and personal characteristics. Furthermore, this information may contribute to the planning and implementation of interventions and public politics at individual and contextual levels, appropriate to reduce barriers and promote functioning of individualswith disabilities.