Trial Outcomes & Findings for Snacks, Smiles and Taste Preferences (NCT NCT03631992)

Of 172 enrolled participants, 140 participants (70 mothers, 70 children) reflecting 70 caregiver-child dyads were randomized.

32 participants (16 mothers, 16 children) or 16 caregiver-child dyads were not randomized for failure to meet inclusion criteria (22 participants; 11 mothers, 11 children), failure to understand study tasks (4 participants; 2 mothers, 2 children), and failure to complete randomization screening tasks (6 participants; 3 mothers, 3 children).

70 participants (35 mothers, 35 children) or 35 caregiver-child dyads were randomized to each arm for a total of 140 participants (70 mothers, 70 children) or 70 caregiver-child dyads; baseline characteristics are reported here for CHILD AGE.by arm (Low Sweet, n=35 children. Regular Sweet n=35 children)

Children's dietary intake of added sugar (tsp/d) was determined from Automated Self-Administered Recall System (ASA24) at baseline, mid-trial, and end of trial.

Caregivers' dietary intake of added sugar (tsp/d) was determined from Automated Self-Administered Recall System (ASA24) at baseline, mid-trial, and end of trial.

Children's indicated liking of two sets of lower sweetness intervention snack foods (Set 1: Strawberry-banana yogurt, apple slices; Set 2: hummus and baby carrots), assessed on separate occasions. Using hedonic face scales, children indicated their like or dislike for each food. Liking data were not obtained during the COVID-19 lockdown. Set 1 foods were discontinued following the lockdown in a modified version of the protocol moved from in person visits to remote assessments. As such, outcome analyses of children's liking of snacks low in sweetness are reported on complete cases (Set 1 foods: n=40 children; Set 2 foods: n=54 children). This analysis shows the count of and percentage of children who liked SET 1 STRAWBERRY-BANANA YOGURT AND APPLE SLICES at the end of the intervention.

Children's indicated liking of two sets of lower sweetness intervention snack foods (Set 1: Strawberry-banana yogurt, apple slices; Set 2: hummus and baby carrots), assessed on separate occasions. Using hedonic face scales, children indicated their like or dislike for each food. Liking data were not obtained during the COVID-19 lockdown. Set 1 foods were discontinued following the lockdown in a modified version of the protocol moved from in person visits to remote assessments. As such, outcome analyses of children's liking of snacks low in sweetness are reported on complete cases (Set 1 foods: n=40 children; Set 2 foods: n=54 children). This analysis shows the number and percentage of children who liked SET 2 HUMMUS AND BABY CARROTS at the end of the intervention.

Children's intake of novel snacks in grams using weighed intake methods; higher intake and consummatory responses indicate greater acceptance. Children's ad libitum intake of two sets of lower sweetness intervention snack foods was assessed on separate occasion (Set 1: Strawberry-banana yogurt, apple slices; Set 2: hummus and baby carrots), a protocol that directly followed assessment of liking. Weighed intakes were used to measure children's intake of each food in grams. Intake data were not obtained during the COVID-19 lockdown. Set 1 foods were discontinued following the lockdown in a modified version of the protocol moved from in person visits to remote assessments. As such, outcome analyses of children's liking of snacks low in sweetness are reported on complete cases (Set 1 foods: n=40 children; Set 2 foods: n=54 children). This analysis presents INTAKE in GRAMS of SET 1 STRAWBERRY-BANANA YOGURT AND APPLE SLICES at the END OF TRIAL.

Children's intake of novel snacks in grams using weighed intake methods; higher intake and consummatory responses indicate greater acceptance. Children's ad libitum intake of two sets of lower sweetness intervention snack foods was assessed on separate occasion (Set 1: Strawberry-banana yogurt, apple slices; Set 2: hummus and baby carrots), a protocol that directly followed assessment of liking. Weighed intakes were used to measure children's intake of each food in grams. Intake data were not obtained during the COVID-19 lockdown. Set 1 foods were discontinued following the lockdown in a modified version of the protocol moved from in person visits to remote assessments. As such, outcome analyses of children's liking of snacks low in sweetness are reported on complete cases (Set 1 foods: n=40 children; Set 2 foods: n=54 children). This analysis presents INTAKE in GRAMS of SET 2 HUMMUS and BABY CARROTS at the END OF TRIAL.

Children's height was measured in cm and weight in kg; these measures were used to calculate child BMI (kg/m2). Child BMI z-scores were calculated with CDC 2000 growth charts using the LMS method, which standardizes a child's BMI relative to a reference population of U.S. children of the same age and sex. The BMI z-score indicates how many standard deviations the child's BMI is above or below the CDC age- and sex-specific average, with higher positive values indicating higher relative body size compared to children of the same age and sex in the CDC reference population. This analysis compares child BMIz by arm at the END OF TRIAL.

Carbon stable isotopes were assessed using ratio mass spectrometry methods using hair samples collected at the start (T0) and end of 4-month intervention (T4).

Mothers will complete the 19-item Caregivers Feeding Styles Questionnaire \[CFSQ\]. Each item is scored from 1 (never) to 5 (always). Mean item scores were used to assess parental demandingness (12 items) and responsiveness (7 items), with a possible range of 1 to 5 for each scale where higher scores indicate higher levels of the construct. This measure was collected and checked at baseline to ensure that there were no differences by group/arm.

Mothers will complete the 26-item Children's Eating Behavior Questionnaire \[CEBQ\]. Each item is scored from 1 (never) to 5 (always) and are averaged and categorized into aspects of child eating (e.g., enjoyment of food, food responsiveness, satiety responsiveness, emotional overeating; emotional undereating, food fussiness); higher numbers reflect more of the behavior. This measure was collected and checked at baseline to ensure that there were no differences by group/arm.

Children's behavioral activation was measured using the Behavioral Activation Scale, a 20-item self-report questionnaire assessing children's sensitivity to reward and approach motivation across 3 subscales: drive (4 items), fun seeking (4 items), and reward responsiveness (5 items). Each question is answered using a 4-point Likert-type scale, with possible scores ranging from 1 (not true) -4 (very true). Scores for each of the three subscales are calculated as the sum of all items on the subscale, with higher scores indicating higher levels of the construct. Possible scores ranges for subscales are as follows: drive (4-16); fun seeking (4-16), and reward responsiveness (5-20). This measure was collected and checked at baseline to ensure that there were no differences by group/arm.

Mothers will complete the 19-item Palatable Eating Motive Scales (PEMS) questionnaire; each item is scored from 1 (almost never/never exhibits behavior) to 5 (almost always/always exhibits behavior). Scores are averaged and categorized to reflect motives for intake of palatable foods (e.g., to socialize, cope, fit in or conform, for reward enhancement); higher numbers reflect more of the motivation.This measure was collected and checked at baseline to ensure that there were no differences by group/arm.