Dose Response of Leucine on Muscle Maintenance During Weight Loss

Bone mass

Bone mineral density

Waist Circumference

Waist circumference to hip circumference ratio

Fat free mass measured using BIA

Skeletal muscle mass measured using BIA

Sum of fat free mass in upper and lower limbs measured using BIA

Visceral fat area measured using BIA

Fat-free mass in kilograms divided by square of height in metres

Skeletal muscle mass in kilograms divided by square of height in metres

Appendicular fat-free mass in kilograms divided by square of height in metres

Brunel Mood Scale (BRUMS). BRUMS is the modified and shorter version of the Profile of Mood Stats (POMS). The 24-item BRUMS was used to assess mood responses. It has six subscales (tension, depression, anger, vigour, fatigue, and confusion) of four mood descriptors each, with respondents rating how they feel "right now" on a 5-point Likert-type scale (0 = not at all, 1 = a little, 2 = moderately, 3 = quite a bit; 4 = extremely). During its development, the BRUMS showed robust psychometric properties across samples of adult students, adult athletes, young athletes, and schoolchildren, and has since demonstrated psychometric integrity for use with many other population groups (doi.org/10.3390/su13116116).For each subscale, the raw score are summed and converted to a T-score based on the norms published in the paper (doi.org/10.3390/su13116116). Higher T-score for each subscale reflects higher profile of that subscale (i.e. tension, depression, anger, vigour, fatigue, or confusion).

Perceived Stress Scale (PSS). The 10-item PSS asks participants about feelings and thoughts during the last month, with responses ranging from 0 (never) to 4 (very often). The final score obtained by reversing responses (e.g., 0 = 4, 1 = 3, 2 = 2, 3 = 1 \& 4 = 0) to the four positively stated items (items 4, 5, 7, \& 8) and then summing across all scale items. Higher scores indicating higher levels of perceived stress.

The Food Craving Questionnaire-Trait Reduced (FCQ-T-r) measures the habitual (unspecified time frame) trait craving for food in general. It predicts food cravings in daily life. It represents a stable trait but is also sensitive to changes in eating behaviour. The questionnaire consists of 15 questions, participants are required to rate the frequency of each statement with a 6-likert scale (never or not applicable = 1, to always = 6). By summing up the scores of the scales, higher scores represent higher level of food cravings.

Food Craving Inventory (FCI) measures food craving for specific foods and the versions validated for use in United States and United Kingdom were used. Questionnaire items pertaining to specific local foods were also added for a combined total of 61 food items (i.e. questionnaire items). For each item, participants were asked how often they have experienced a craving for that particular food on a 5-point Likert scale (never = 1, to always/almost every day = 5). The higher the score, the higher the food craving.

Three-Factor Eating Questionnaire-R18 - cognitive restraint component (TFEQ-r18 Cog). It has 6 items and each scored 1 to 4 for a total of 24 points. The higher the total score, the more disordered the eating behaviour.

Hunger over past 4 weeks. Visual analogue scale 0 to 100. The higher the number the greater the amount of hunger.

Baseline appetite using visual analogue scale 0 to 100. The higher the score, the higher the baseline appetite

Satiety over past 4 weeks. Visual analogue scale 0 to 100. The higher the number, the greater amount of satiety.

Diet satisfaction over past 4 weeks. Visual analogue scale 0 to 100. The higher the number the greater the satisfaction.

Implicit satiety goal (unconscious, habitual, or subjective levels of fullness that individuals aim to achieve when eating) using visual analogue scale 0 to 100. The higher the number, the greater implicit satiety goal.

Ideal portion size (IPS). Participants were presented with a diverse range of 25 foods typically consumed in Singapore. The computerised portion selection tasks presented a series of 50 images of each food that increased in portion size in 20 kcal increments (20-1000 kcal). Participants could increase or decrease the portion size by 20 kcal using the left and right arrow-keys on the keyboard and were encouraged to view the full range of portions before selecting their ideal portion. Foods were presented in a randomised order and an average portion size for each participant was calculated by averaging the selected portion sizes of all 25 foods (α = 0.95). This created an overall index of IPS.

Expected satiety (ES) measured using computerised task. The ES task followed the same format as the IPS task, in other words, 50 images were presented depicting 20 kcal increment changes in portion size. This task sought to measure participants' expectations about how satiating a diverse range of foods were. Larger portions selected indicated that participants require larger quantities of the food to delay the onset of hunger, in other words they have lower expected satiety. The same 25 foods presented in the IPS were also presented in the ES in a randomised order and were averaged to create an overall index of ES (α = 0.95).

A single image of each food item was presented to participants alongside questions of food-related characteristics. Participants were asked to rate how filling they expected each food to be (not filling at all (0) to very filling (100)), how much they like each food in general (not at all (0) to very much (100)) and how often they eat each food (never (0) to more than once a week (5)). Each food was presented in a randomised order and average scores across all 25 foods for participants perceptions of how filling, frequently consumed and preference for foods were calculated.