Page 156 - GHES-2-1

Global Health Econ Sustain Total sugar consumption Philippines

“Visuals of Foods,” containing pictures of food items, was total mean sugar intake of each age group (Centers for
used to aid respondents in correctly identifying the food Disease Control and Prevention, n.d.).
consumed. A Food Composition Library was subsequently Data on consumption of different food categories were
used to estimate the nutrient contents of consumed food. obtained from the 2013 National Nutrition Survey. The
For foods that were not found in the Food Composition change in consumption of each food category from 2008
Library, a “List of Alternates and Substitutes” was used to to 2013 was calculated and used as a proxy measure to
determine substitutes for foods consumed by respondents. determine the trend in sugar intake from different sources

2.3. Data analysis during the same period. The t-test was used to determine
significant changes in food category intake.
The Philippine Food Composition Table does not provide
nutrient values for total sugar. Hence, the sugar content of 2.4. Grouping of foods
all foods consumed was estimated from total sugar values The Food and Agriculture Organization (FAO) (2015)
derived from other food composition tables, using the recommended that the level of food processing should be
process described by INFOODS. The INFOODS guidelines taken into account when examining food consumption
for food matching (FAO/INFOODS, 2012) were used to data to inform the development and implementation of
guide the selection of appropriate foods from which to food-based guidelines and approaches for the prevention
borrow values for total sugar, in the most appropriate source of chronic diseases. The NOVA food classification system
of compositional data. Total sugar, rather than free sugar, was (Fardet et al., 2015), developed by researchers in Brazil,
estimated since food composition data do not distinguish classifies food according to the nature, degree, and purpose
free sugars from total sugar. Briefly, the INFOODS guidelines of processing. The present study used a modified version of
recommend the following steps: (i) provide sufficient food the NOVA classification, wherein foods were classified into
identification using descriptors and taxonomic/scientific two groups: (i) minimally processed foods (comprising
names of the food to allow unambiguous identification; and cooked whole foods, e.g., boiled rice and tubers, whole
(ii) if the water content between two foods differed by more fish, meat, and chicken dishes, milk [fresh liquid and
than 10%, the estimation of nutrients (in this case, total whole milk powder], and raw or cooked whole vegetables
sugar) should be adjusted accordingly. Sugar consumption and fruits), and (ii) processed foods and food products
was then computed by multiplying the total sugar content of (comprising processed and preserved food products, i.e.,
each food by the amount of food ingested by each individual foods made from processed ingredients).
and subsequently adding up the total.
Most foods consumed in the Filipino diet are processed
Usual sugar intake was estimated using the software or cooked to a certain extent before ingestion (i.e., raw
for intake distribution estimation (PC-SIDE) software foods are not usually eaten except for fruits). For purposes
(Iowa State University, 2018). PC-SIDE implements the of this study, minimally processed foods are defined as
method developed by Nusser et al. (1996) to estimate the those foods whose traditional cooking and preparation
distribution of the usual intake of nutrients and foods. The processes do not include sugar as part of the recipe.
method adjusts dietary intake data by shifting the observed Therefore, minimally processed foods were assumed to
intake data away from zero and uses a regression-based contain mostly natural sugar and have little or no added
ratio adjustment to transform the dietary intake data into sugar. Processed foods are commercially sold food
normality. This is followed by an estimation of within products. Due to processing, these foods are considered to
and between individual variances for the intakes, and the contain mostly added sugar rather than natural sugar.
intakes are then transformed back into the original scale.
In summary, all foods consumed by survey respondents
Percentiles and interquartile range (IQR) of sugar
intake were estimated using STATA. The proportion of were listed, and similar foods were grouped into specific
categories (18 categories were created for 1306 individual
energy from sugar was estimated by initially calculating food items consumed in 2008). Each food category was
calories from sugar (i.e., multiplying the usual sugar intake classified into one of these two groups that correspond to
[in g] by 4 kcal/g [or 4.184 kJ/g]). The proportion of energy the dietary patterns of this specific population (Table 1).
from sugar consumed by the different age groups was then
calculated by dividing the kcal of total sugar intake in each 3. Results
group by the total energy intake (ratio of means) (Centers
for Disease Control and Prevention, n.d.). To identify 3.1. Characteristics of the study sample
sources of sugar, the proportion contributed by each food The characteristics of the study subjects are displayed in
category to usual intake was calculated, wherein the total Table 2. Most of the subjects were males from rural areas
mean sugar intake from each category was divided by the and lower wealth quintiles. In addition, stunting and

Volume 2 Issue 1 (2024) 3 https://doi.org/10.36922/ghes.2060