Measures

At baseline, trained social workers administered a questionnaire to collect maternal sociodemographic characteristics. We used the Mexican Association of Marketing Research and Public Opinion Index (AMAI), a 13 × 6 battery based on family possessions, living conditions, and education, to categorize six socioeconomic status (SES) levels ranging from lowest (1) to highest (6) SES within the cohort [22]. Based on the exploratory analysis of its pattern association with weight trajectories, we reclassified these categories as low (1, 2), medium (3), and high (4, 5, 6). Women self-reported their pre-pregnancy weight (kg) six months before pregnancy. We adjusted these self-reported weights using an imputation procedure to account for potential recall bias [23]. Maternal pre-pregnancy weight was predicted using a longitudinal model (linear mixed-effects) with repeated weight data from six months before pregnancy to the third-trimester visit. We used tenfold cross-validation to test the model’s performance based on those women with weights (n = 87) close to their last menstrual period (LMP). The root mean squared error of 3.21 kg was considered a measure of predictive accuracy [24]. We calculated pre-pregnancy BMI using adjusted pre-pregnancy weight (kg) and measured height (m) at the baseline visit. Pre-pregnancy BMI was classified using World Health Organization criteria (WHO, 2005): underweight (< 18.5 kg/m²), healthy weight (18.5–24.9 kg/m²), overweight (25.0–29.9 kg/m²), and obesity (≥ 30 kg/m²).

Trained personnel measured women’s weight (without shoes and clothes) during pregnancy (baseline and third-trimester visits) and postpartum (one-, six-, and 12-month visits). Weight was measured using a combined mechanical scale and stadiometer (Health-O-Meter; Scaleomatics INC, Cleveland, OH) to the nearest 100 g. Following the Institute of Medicine’s (IOM, 2009) recommendations for gestational weight gain rate [7], we classified women’s gestational weight gain at the second and third trimesters of pregnancy (considering pre-pregnancy BMI and gestational age) as insufficient, adequate, and excessive. Gestational age was defined according to the LMP and corroborated with the Capurro method. We adjusted gestational age weeks in 33 cases where gestational age at birth differed by more than two weeks with LMP.

At each study visit (except 12 months postpartum), women reported the average daily time spent in physical activities, such as walking, and sedentary behaviors, such as watching television or reading. Physical activity was categorized as inactive (< 20 min per day), minimally active (20 – 60 min per day), and active (> 60 min per day). Sedentary behavior was categorized as non-sedentary (< 120 min per day), moderate sedentary (120–180 min per day), and sedentary (> 180 min per day) [25]. At one and six months postpartum, women reported breastfeeding their infant (yes or no) and if they were exclusively providing breast milk (yes or no). Breastfeeding was categorized as not breastfeeding, not exclusive breastfeeding, and exclusive breastfeeding. The intensity of breastfeeding was defined as the daily number of times a woman breastfed her baby at one and six months postpartum.