Bibian van der Voorn

80 CHAPTER 6 Study 1. Stepwise GEEs were performed. First, the influence of very preterm delivery (as a dichotomous factor) onmilk cortisol concentration was investigated. Second, the influence of time of collection (as a continuous factor) on milk cortisol concentration was investigated. Third, the influence of very preterm delivery on milk cortisol concentration, adjusted for time of collection, was investigated. Analyses were repeated for milk cortisone concentration and the cortisone-to-(cortisol + cortisone) ratio. Study 2. Stepwise GEEs were performed to assess the diurnal pattern of milk cortisol concentrations. First, the influence of time of collection (as a continuous factor) on milk cortisol concentration over time was assessed in a linear model. Second, linear modeling was repeated after dividing the day into 2 parts on the basis of visual peak identification: the increase (i.e., 0000 to 0700) and the decline (i.e., 0701 to 2359). Third, the influence of time of collection (as a continuous factor) on milk cortisol concentration over time was assessed in a quadratic model (i.e., y = a + bx + cx 2 , with y representing milk cortisol concentration and x the time of collection). Fourth, the influence of time of collection (as a continuous factor) on milk cortisol concentration over time was assessed in a cubic model (i.e., y = a + bx + cx 2 + dx 3 ). Analyses were repeated for milk cortisone concentration and the cortisone-to-(cortisol + cortisone) ratio. Associations between milk and salivary concentrations were tested by using Pearson correlation and linear regression analysis. RESULTS Study 1. Table 1 presents baseline characteristics of the mothers who delivered very preterm compared with those who delivered full-term infants. All of the children had an appropriate size for GA. TABLE 1. Characteristics of participants: Study 1. Mothers who delivered Full-term Mothers who delivered Preterm Age Mother, yr 33 (29 to 39) 33 (24 to 40) Parity 2 (1 to 4) 1 (1 to 3) Vaginal delivery 6 (60 %) 7 (70 %) Gestational age, wks 39.0* (37.6 to 41.3) 28.7 (24.7 to 30.9) Birth weight, g 3392* (2155 to 4064) 1058 (660 to 1845) Glucocorticoid treatment for impending preterm 1* (10 %) 8 (80 %) Values are median (ranges) or frequencies (%), n= 10 vs. 10 * P<0.05 between groups.