ABSTRACT
O estudo propôs verificar a autenticidade científica do Instrumento de Avaliação do Nível de Atividade Física de Idosos (INAFI). Participaram 64 idosos, com 69,37 ± 6,70 anos (mulheres) e 70,5 ± 3,54 anos (homens). O instrumento adotado foi o acelerômetro GT3X, Actigraph (USA). Na análise geral do instrumento, comparação entre os grupos A e B (reprodutibilidade, mesmo entrevistador), o INAFI apresentou bom desempenho, p = 0,6956, grupos B e C (objetividade, entrevistadores diferentes), o instrumento foi razoável, p = 0,5047. No diagrama de dispersão observou-se que o viés A e B (= 0,5), B e C (= 6,61), é muito próximo de Zero, demonstrando que há concordância na maneira de aplicar as entrevistas. Na comparação, INAFI X acelerômetro foi necessário homogeneizar a diferença no tamanho das escalas, dividindo as variáveis pelas suas respectivas médias. Os resultados das análises: total/INAFI (MET/min/sem) X counts/min eixo1; total/INAFI (MET/min/sem) X counts/min vetor magnitude; total/atividade leve/INAFI, (< 3 METs) X total/light e lifestyle/acelerômetro (1,5-2,9 METs); total/atividade moderada/INAFI (> 3 METs) X total/atividade moderada/acelerômetro (3-5,9 METs) obteve-se um viés (= zero), portanto existe uma forte concordância entre as medidas dos dois instrumentos. Podemos afirmar que o INAFI se mostrou eficiente ao propósito para qual foi construído, portanto possui autenticidade científica. (AU)
The study proposed to verify the scientific authenticity of the Instrument Measuring the Level of Physical Activity in the Elderly (INAFI). Sixty-four elderly individuals participated, with 69.37 ± 6.70 years (women) and 70.5 ± 3.54 years (men). The instrument adopted was the GT3X accelerometer, Actigraph (USA). In the general analysis of the instrument, comparison between groups A and B (reproducibility, same interviewer), INAFI presented good performance, p = 0.6956, groups B and C (objectivity, different interviewers), the instrument was reasonable, p = 0.5047. In the dispersion diagram it was observed that the bias A and B (= 0.5), B and C (= 6.61), is very close to Zero, demonstrating that there is agreement in the way of applying the interviews. In the comparison, INAFI X accelerometer was necessary to homogenize the difference in the size of the scales, dividing the variables by their respective means. The results of the analyzes: total / INAFI (MET / min / without) X counts / min axis1; total / INAFI (MET / min / without) X counts / min vector magnitude; total / light activity / INAFI, (<3 METs) X total / light and lifestyle / accelerometer (1.5-2.9 METs); total / moderate activity / INAFI (> 3 METs) total X / moderate activity / accelerometer (3-5.9 METs) was obtained a bias (= zero), therefore there is a strong agreement between the measurements of the two instruments. We can say that INAFI was efficient for the purpose for which it was built, so it has scientific authenticity. (AU)