Opaque Bottle Study

During post-tests, participants in both the Clear Bottle and Opaque Bottle Groups were asked to use the clear bottle during one feeding observation and the opaque bottle during the other feeding across the two post-test assessments, which ensured research assistants remained masked to whether the participant was in the intervention or control group and allowed for within-subject comparison of effects of clear versus opaque bottles on maternal sensitivity and infant intake; order of bottle presentation was counterbalanced. For both opaque and clear bottle-feedings, infant intake was assessed by weighing the infant's bottle before and after each feeding observation using a top-loading balance (SP601 Scout Pro Portable Balance, Ohaus, New Jersey, USA).

Time frame: 12-week period

Population: Overall Number of Participants Analyzed for each group (n=30) is not consistent with the number of participants indicated in the Participant Flow module because data were analyzed at the level of the dyad (n = 30 dyads per group), not at the level of each individual member of the dyad (n = 60 individuals per group, which represents 30 mothers + 30 infants)

Infants' weight and length were assessed during home-based assessments. Infant weight and length were measured in triplicate using a portable infant scale/infantometer. To minimize bias, infants were always weighed and measured while only wearing a clean diaper and prior to the observed feeding. Weight and length values were standardized to age- and sex-specific weight-for-length z-scores based on the World Health Organization Child Growth Standards (https://www.who.int/tools/child-growth-standards). A z-score of 0 indicates that the child's weight status is at the population median; standard deviations closer to 0 represent healthier weight status. Standard deviations below 0 indicate the child's weight status is below the median, with values \< -2 indicating the child is underweight. Standard deviations above 0 indicate the child's weight status is above the median, with values \> 2 indicating the child is overweight or obese.

Time frame: 12-week period

Population: Overall Number of Participants Analyzed for each group is not consistent with the number of participants indicated in the Participant Flow module because data were analyzed at the level of the dyad (n = 30 dyads per group), not at the level of each individual member of the dyad (n = 60 individuals per group, which represents 30 mothers + 30 infants). In addition, weight-for-length data were missing for n=12 dyads in the Clear Bottle Group and n=13 dyads in the Opaque Bottle Group.

Mothers' behavior during feeding observations was coded using the Nursing Child Assessment Parent-Child Interaction - Feeding Scale (NCAFS). This scale contains six subscales, four of which describe maternal attributes (Sensitivity to Cues, Response to Child's Distress, Social-Emotional Growth Fostering, and Cognitive Growth Fostering) and two of which describe infant attributes (Clarity of Cues and Responsiveness to Caregiver). The present analysis focused on the Sensitivity to Cues subscale, which measures the degree to which the mother is able to understand and respond to her child's cues. This scale provides a global measure of how sensitive the mother is to the infant's needs during the feeding interaction. Possible score range is 0-16, which is derived from summing together maternal scores for each of the 16 subscale items. Higher scores indicate greater sensitivity to infant cues.

Time frame: 12-week period

Population: Overall Number of Participants Analyzed for each group (n=30) is not consistent with the number of participants indicated in the Participant Flow module because data were analyzed at the level of the dyad (n = 30 dyads per group), not at the level of each individual member of the dyad (n = 60 individuals per group, which represents 30 mothers + 30 infants)

Subscapular skinfold thickness, a marker of infant adiposity, was assessed during home-based assessments by trained research assistants. A study-specific Harpenden caliper was used for this assessment. Subscapular skinfold thickness values were standardized to age- and sex-specific z-scores based on the World Health Organization Child Growth Standards (https://www.who.int/tools/child-growth-standards). A z-score of 0 indicates that the child's adiposity is at the population median. Standard deviations below 0 indicate the child's adiposity is below the median and standard deviations above 0 indicate the child's adiposity is above the median. Standard deviations closer to 0 represent healthier levels of adiposity.

Time frame: 12-week period

Population: Overall Number of Participants Analyzed for each group is not consistent with the number of participants indicated in the Participant Flow module because data were analyzed at the level of the dyad (n = 30 dyads per group), not at the level of each individual member of the dyad (n = 60 individuals per group, which represents 30 mothers + 30 infants). In addition, skinfold data could not be obtained for n=23 dyads in the Clear Bottle Group and n=23 dyads in the Opaque Bottle Group.

Triceps skinfold thickness, a marker of infant adiposity, was assessed during home-based assessments by trained research assistants. A study-specific Harpenden caliper was used for this assessment. Triceps skinfold thickness values were standardized to age- and sex-specific z-scores based on the World Health Organization Child Growth Standards (https://www.who.int/tools/child-growth-standards). A z-score of 0 indicates that the child's adiposity is at the population median. Standard deviations below 0 indicate the child's adiposity is below the median and standard deviations above 0 indicate the child's adiposity is above the median. Standard deviations closer to 0 represent healthier levels of adiposity.

Time frame: 12-week period

Population: Overall Number of Participants Analyzed for each group is not consistent with the number of participants indicated in the Participant Flow module because data were analyzed at the level of the dyad (n = 30 dyads per group), not at the level of each individual member of the dyad (n = 60 individuals per group, which represents 30 mothers + 30 infants). In addition, skinfold data could not be obtained for n=23 dyads in the Clear Bottle Group and n=23 dyads in the Opaque Bottle Group.

Infant central adiposity was assessed via a waist circumference measurement, which was assessed during home-based assessments by trained research assistants. Waist circumference was assessed in triplicate using a study-specific Gulick II tape measure.

Time frame: 12-week period

Population: Overall Number of Participants Analyzed for each group is not consistent with the number of participants indicated in the Participant Flow module because data were analyzed at the level of the dyad (n = 30 dyads per group), not at the level of each individual member of the dyad (n = 60 individuals per group, which represents 30 mothers + 30 infants). In addition, waist circumference data could not be obtained for n=16 dyads in the Clear Bottle Group and n=16 dyads in the Opaque Bottle Group.

Mothers' perceptions of their assigned bottles, including any perceived barriers to their use, were assessed via a questionnaire developed by the investigators.

Time frame: 12-week period

Population: Overall Number of Participants Analyzed for each group (n=30) is not consistent with the number of participants indicated in the Participant Flow module because data were analyzed at the level of the dyad (n = 30 dyads per group), not at the level of each individual member of the dyad (n = 60 individuals per group, which represents 30 mothers + 30 infants)