Trajetórias de atividade física medida por acelerometria nos primeiros anos de vida e seus determinantes e correlatos: estudo na Coorte de Nascimentos de 2015 de Pelotas (RS)

Abstract

The objective was to describe the trajectories of physical activity (PA) from one to four years old of children belonging to the 2015 Birth Cohort of Pelotas (RS) and to verify the associations with sociodemographic, gestational, perinatal, environmental, maternal, behavioral, and breastfeeding determinants and correlates. The study used data from the perinatal, one-, two-, and four-years follow-ups. PA was measured using accelerometers worn on the left wrist for two days at one and two years, and seven days at four years and the data were used in raw form in milli-g. One- to four-year PA trajectories were plotted for children with three measures of PA (n=1,798) using group based trajectory modeling. Exposures were collected by questionnaire and Poisson regression models with robust variance were used to test associations with trajectories. In article II, associations of correlates with PA at 1 (n=2,974), 2 (n=2,645), and 4 years (n=2,955) were tested by linear regression. Two PA trajectories (Moderate and High) were identified, both showing a linear increase in PA in the first years, differing in volume. In article I, female sex (PR: 0.91; 0.88/0.94), high maternal education (PR: 0.93; 0.88/0.97), and high birth weight (PR: 0.91; 0.85/0.97) reduced the probability that the child would be in the High PA, and birth order ≥3 (PR: 1.06; 1.01/1.11) increased this probability. In article II, attending school in some follow-up (PR: 0.95; 0.91/0.99) and always attending public school (PR: 0.95; 0.90/0.99) reduced the probability of the child being in the High PA. Children of mothers who performed any leisure PA (PR: 1.08; 1.01/1.15) and other leisure PA (except walking) in more than one follow-up (PR: 1.13; 1.04/ 1.21) and children who had adequate sleep time at 2 (PR: 1.11; 1.02/1.21) or <2 follow-ups (PR: 1.10; 1.01/1.21) were more prone to the High PA trajectory. At one year, partial breastfeeding (β: 0.56; 0.08/1.05) was positively associated with PA and excessive screen time, negatively (β: -1.24; -1.77/-0,72). At two years, attending public (β: -1.71; -3.07/-0.35) or private school (β -1.32; -2.46/- 0.17) were negatively associated with PA, while the maternal leisure walk (β: 1.47; 0.11/2.83) and the mother perceived her child as more active than children of the same age (β: 1.69; 0.25/3.13) were positively associated with childhood PA. At four years, duration of breastfeeding for more than six months (except children who continued to be partially breastfed), ≥2 siblings (β: 1.58; 0.27/2.89), other maternal leisure PA except walking (β: 1.94; 0.64/3.24) and the mother perceived her child as more active than children of the same age (β: 3.59; 2.61/4.56) were positively associated with PA levels. Public school (β: -1.79; -2.95/-0.65) and the mother perceived the child as less active (β: -4.80; -6.93/-2.67) were negatively associated with PA. These characteristics should be considered when planning PA programs and interventions for young children.