Skip to main content
. Author manuscript; available in PMC: 2023 Feb 1.
Published in final edited form as: Eur J Nutr. 2021 Jul 3;61(1):127–140. doi: 10.1007/s00394-021-02616-x

Table 2:

Linear mixed-effects regression models examining urine osmolality by week of pregnancy for the full sample and stratified by intervention status

Full Sample Full Sample Control Group Intervention Group

Independent variables Model 1.1 Urine osmolality (mOsm/kg) (SE) Model 1.2a Urine osmolality (mOsm/kg) (SE) Model 1.3 Urine osmolality (mOsm/kg) (SE) Model 1.4 Urine osmolality (mOsm/kg) (SE)
Week of pregnancy 28.7*** (6.7) 39.8*** (8.5) 39.2*** (9.4) −1.4 (7.9)
(Week of pregnancy)2 −0.6*** (0.1) −1.0*** (0.2) −0.9*** (0.2) −0.1 (0.2)
Intervention x week of pregnancy interaction −22.5** (9.7) −41.3***(12.3) -- --
Intervention x (week of pregnancy)2 interaction 0.5** (0.2) 1.0*** (0.3) -- --
Intervention 206.1* (116.7) 384.0*** (142.3) -- --
Pre-pregnancy BMI (kg/m2) 5.2 (3.9) 7.3* (4.4) −2.3 (5.4) 18.9*** (5.9)
Maternal age (years) −2.9 (9.3) 2.6 (9.4) 2.5 (10.8) −2.6 (12.6)
Steps per day (per 1000) -- −2.8 (2.5) −4.4 (3.9) −1.0 (3.2)
Average weekly weight change from pre-pregnancy weight (lb) -- 2.3 −2.2 9.3***
Total energy intake (100 kcal) -- −0.3 (1.5) −0.5 (2.3) 0.7 (1.9)
Observations b 653 413 199 214
Number of participants 27 25 13 12
***

p<0.01

**

p<0.05

*

p<0.1

Note: 3-level mixed-effects model, with random effects (intercepts) for each woman nested within a random effect for intervention status, and within participants. 2-level mixed-effects model for Models 1.3 (control group) and 1.4 (intervention group) since model is stratified by intervention status. Results for Models 1.1 and 1.2 indicate the average change in Uosm in relation to week of pregnancy and in relation to the intervention, testing for an interaction between week of pregnancy and intervention group, allowing for quadratic relationships with week of pregnancy, and controlling for the covariates listed. Models 1.3 and 1.4 indicate the average change in Uosm within the control and intervention groups, respectively, in relation to each additional week of pregnancy, allowing for quadratic relationships with week of pregnancy, and controlling for the covariates listed.

a

Missing 2 women that did not have both weight and steps data on any sampled weeks.

b

Changes in number of observations across models due to missing weekly steps and weight data.