Comparação das variáveis de atividade física fornecidas pelo acelerômetroactigraph GT3X e pelo aplicativo de celular google fit durante a marcha de indivíduos pós-acidente vascular encefálico

Abstract

The objective evaluation of physical activity levels of individuals with stroke becomes very important for clinicians involved in stroke rehabilitation, once it guides the professionals to set more realistic and objective goals to improve physical conditioning of these individuals. In this scenario, the use of accelerometry and smartphone applications stands out, since theyprovide objective measures of different physical activity variables, such as the number of steps taken and energy expenditure (EE). However, although these devices have been frequently used in recent studies with individuals with stroke, it is not known if their data represent the actual physical activity levels of these individuals. Therefore, in the present dissertation, two studies were carried-outin an attempt to solve these issues. The first study aimed at comparing the number of steps predicted by the ActiGraph GT3X accelerometer and the Google Fit smartphone application, with the number of steps observed by the researcher during fast overground walking of chronic stroke individuals. The second study aimed at comparing the EE estimates from the ActiGraph GT3X accelerometer and the Google Fit smartphone application, with the EE obtained from the conversion of the oxygen consumption (VO2) given by the Metamax 3B ergoespirometer during fast overground walking of chronic stroke individuals. Both studies had a cross-sectional design, in which individuals with chronic stroke were asked to walk on a 10-meter straight hallway over five minutes attheir fast speeds, wearing the ActiGraph GT3X accelerometer, a smartphone containing the Google Fit application, and the Cortex Metamax 3B ergoespirometer. The criterion-standard measure for the variable related to the number of steps was thatcounted by a trained examiner. The inclusion criteria were: ages 20 years, time since stroke onset >6 six months, ability to walk at least 14m independently, ability to understand and follow verbal instructions, and absence of cognitive deficits, as determined by the cut-off scores on the Mini Mental State Exam. Individuals, who had any other neurological, orthopedic, and/or respiratory diseases, were excluded. Descriptive statistics, normality tests (Shapiro-Wilk) were carried-out for all outcomes, followed by thecalculation of Pearson's correlation coefficients and intra-class correlation coefficient (ICC [2.1]). For all analyses, the significance level was established at 0.05. Thirty-seven individuals were included in the present study, who had a mean age of 62 (±11.2) years, and a mean time since the stroke onset of 91.3 (±90.4) months. Significant and positive associations were found between the number of steps observed by the researcher and the number of steps estimated by the Google Fit smartphone application (r=0.89, p<0.001), and the ActiGraph GT3X accelerometer (r=0.56; p<0.001). The ICC (2,1) analysis revealed thatthe Google Fit smartphone application showed greater agreement (ICC=0.93; p <0.001) and a lower mean difference between the observed and estimated number of steps (p=0.37), whereas the ActiGraph GT3X accelerometer data showed lower agreement (CCI=0.32, p<0.001) and a mean difference between the observed and estimated number of steps of 191.8 (p < 0.001) steps. Regarding the EE, significant, weak, and positive association was only found between the EE estimated from the combined formula from ActiGraph GT3X and that converted from the ergospirometer (r=0.37; p=0.04). The ICC analyses (2,1) found no agreement between these EE data. Therefore, the results of the present study demonstrated that, despite being frequently used in studies withstroke individuals, the ActiGraph GT3X accelerometer did not provide validmeasures, and maynot be the most appropriate physical activity monitor for this population, since its variables did not show any association with the criterionstandard measure. On the other hand, the Google Fit smartphone application showed the potential to be used with individuals with chronic stroke, since the number of steps estimated by the device was associated with the criterionstandard measure during fast overground walking.