The study sample was comprised of 04-05-year-old preschool children; the majority (70%) were four years old (48-59-months). An equal number of girls and boys participated in this study.
Amount of free sugar intake
As shown in Table 1, the mean (SD) and median (IQR) amount of free sugar intake were 79 (68.6) g/day and 57.9(33.2- 95.8) g/day (equivalent approximately to 14.5 (8.3-23.9) teaspoons/day), respectively. Free sugar contributed 21.1% (12.5%-35.9%) of total energy intake on average. The median (IQR) composite frequency of sugar intake was four times (2.7-6.1) per day.
Table 1
The amount and of free sugar intake among the participants (N= 813)
|
Sugar energy* as a percentage of total energy requirement**
|
Amount of sugar intake (g/day)
|
Frequency of sugar intake (times/ day)
|
Range
|
0.5-192.3
|
2.5 - 520.0
|
0 - 15.8
|
Mean (SD)
|
29.0 (25.3)
|
79.0 (68.6)
|
4.9 (3.2)
|
Median (IQR)
|
21.1 (12.5 - 35.9)
|
57.9 (33.2 - 95.8)
|
4.0 (2.7 - 6.1)
|
*Sugar energy= amount of sugar (g) x 4 Kcal/ g (1g of sugar provides 4kCal of energy)
**Total energy requirement= bodyweight x energy requirement/ kg body weight/ day
***IQ range = Inter-Quartile Range
The percentage contribution of different sugary food and beverage groups to total free sugar intake
The contribution of each food and beverage group to total free sugar intake is shown in Figure 1. To emphasize excess consumption, biscuits were analysed separately from the other bakery products.
The main contributors to the total intake of free sugar were bakery products (28%), followed by biscuits (15%), and table sugar (15%).
Figure 2 shows the percentage of participants who consumed each food category at least once a month.
The largest percentage of children consumed desserts (96.8%), followed by biscuits (95.6%) and bakery products (95.6%).
Factors associated with the amount of free sugar intake
Multivariate logistic regression with backward elimination was performed after bivariate analysis to identify factors associated with high free sugar intake. As shown in Table 2, the final model retained six predictor variables.
R2 value for the final model was 0.215, indicating a 21.5% variance in the total sugar intake. This model showed an F value of 25.662 with a statistically significant p-value (p< 0.01).
Table 2
An analysis of multiple logistic regression models relating total free sugar intake to the factors associated with it
|
|
Unstandardized Coefficients
|
Standardized Coefficients
Beta
|
t
|
Sig.
(p)
|
95% Confidence Interval for B
|
|
|
B
|
Std. Error
|
Lower Bound
|
Upper Bound
|
|
(Constant)
|
69.27
|
15.979
|
|
4.335
|
<0.001
|
37.8
|
100.658
|
- Ethnicity
Sinhala
Other ethnic groups*
|
17.27
|
6.882
|
0.09
|
2.51
|
0.012
|
3.7
|
30.8
|
- Maternal education level
Up to GCE O/L** bellow
Up to GCE A/L*** above
|
-9.82
|
4.792
|
-0.08
|
-2.04
|
0.041
|
-19.2
|
-0.4
|
- School-going siblings
No
Yes
|
10.39
|
4.754
|
0.08
|
2.18
|
0.029
|
1.1
|
19.7
|
- Eating while returning from preschool
No
Yes
|
38.46
|
5.245
|
0.28
|
7.33
|
<0.001
|
28.2
|
48.7
|
- Eating while watching TV
No
Yes
|
47.47
|
7.269
|
0.25
|
6.53
|
<0.001
|
33.2
|
61.7
|
- Dental clinic attendance
No
Yes
|
-19.2
|
5.093
|
-0.14
|
-3.78
|
<0.001
|
-29.3
|
-9.3
|
*Non-Sinhala- Tamil, Moor and others amalgamated
**GCE O/L- General Certificate of Education Ordinary Level (Grade 11)
**GCE A/L- General Certificate of Education Advanced Level (Grade 13)
- Predictors- Ethnicity, Religion, Maternal education level, Presence of school-going siblings, eating while coming back from preschool, eating while watching TV, Visited Dental clinics
- Dependent variable: Total free sugar intake
R square= 0.215 Adjusted R Square= 0.207
Maternal educational level and dental clinic attendance were significantly and negatively associated with free sugar intake. A child whose mother had obtained at least G.C.E. (advanced level) education was likely to consume significantly less sugar than a child whose mother had obtained up to G.C.E. (ordinary level) (p=0.04). Children who visited a dental clinic at least once had a significantly lower sugar intake than those who had never visited a dental clinic (p< 0.001).
Ethnicity other than Sinhala (p=0.01), the presence of school-going siblings (p=0.02), eating habits such as eating while returning from preschool (p<0.001), and eating while watching television (p<0.001) were significantly associated with high intake levels of sugar.
As shown in Table 3, free sugar intake from particular groups of sugary foods was significantly related to the regular presence of the same group of foods at home.
Table 3
The relationship between free sugar intake from different groups of sugary foods and the presence of the same group of foods at home on a regular basis
|
Number
|
Mean
|
Median
|
Significance
|
Biscuits
|
|
Sugar from biscuits
|
|
Yes
|
698
|
12.1
|
5.8
|
z=3.77
|
No
|
115
|
8.3
|
5.6
|
p<0.001
|
Sugar sweetened beverages
|
|
Sugar from sugar sweetened beverages
|
|
Yes
|
182
|
8.3
|
3.7
|
z=3.55
|
No
|
631
|
5.6
|
2.3
|
p<0.001
|
Chocolate confectionery
|
|
Sugar from chocolate confectionery
|
|
Yes
|
201
|
6.1
|
2.7
|
z=3.77
|
No
|
612
|
4.3
|
1.4
|
p<0.001
|
Fruits
|
|
Total free sugar
|
|
Yes
|
703
|
73.9
|
53.2
|
z= 1.99
|
No
|
110
|
83.3
|
66.3
|
p=0.04
|
Total
|
813
|
|
|
|
Among those who regularly have biscuits (p<0.001), beverages (p<0.001), and chocolates (p<0.001) at home, sugar intake from relevant items was significantly higher than that of their counterparts. The opposite pattern was observed among children for whom fruits were available at home on a regular basis, who had a significantly lower intake of total free sugar than those who did not (p=0.04).