Abstract
Backgrounds and aim: Unhealthy eating behaviours increase with age and have been associated with adverse health consequences in adulthood. We examined the influence of screen-based sedentary behaviours (SBs) on unhealthy food consumption, such as energy-dense foods and sweetened drinks, among a representative sample of nearly 60 000 adolescents and assessed the role of possible modifiers. Methods: Data come from the Italian 2009–10 Health Behaviour in School-aged Children (HBSC) survey. Data on Eating patterns, SBs, physical activity, peers network, BMI and socio-economic status (SES) were collected following the HBSC study protocol. Hierarchical logistic regression models were used. Results: Unhealthy food consumption was significantly associated with a lower intake of fruit and vegetables and with the increase of SBs in both sexes and in all ages. The risk was interestingly higher in normal weight adolescents, in those with wider relationships with peers and in low SES children. Conclusions: This study adds evidence to support the importance of investing more resources in educational initiatives both to increase parents’ awareness to support adolescents on dietary choices and on time spent in screen-based behaviours, independently of their adiposity status; and to develop youth’s ability to access and appropriately use media and technologies. Policy makers should also increase their attention on introducing regulatory policies on television food advertising to which youth are exposed.
Introduction
Childhood nutrition has gained increasing attention in recent years, 1 particularly in relation to the global obesity crisis. Dietary factors, such as consumption of fast food and sugar-sweetened drinks, 2 and the type of dietary fat, rather than total fat consumption, are linked to several undesirable health outcomes. These include type-2 diabetes, hypertension, hypercholesterolemia and other metabolic disturbances, as well as overweight and obesity. 1,3 In particular the intake of partially hydrogenated (trans) fat, commonly found in commercial bakery products and fast foods, may raise the risk for both cardiovascular diseases and type-2 diabetes by increasing the serum levels of low-density lipoprotein cholesterol and decreasing high-density lipoprotein cholesterol. 3 In contrast, unsaturated fats from vegetables can lower the risk of these diseases by favouring an high-density lipoprotein increase. 4
Scholars showed that the 30% of the calories in the average diet of US children derive from sweets, soft drinks, salty snacks and fast food, and consumption of these foods has doubled in the last 30 years. 5 Other studies have reported that one-fifth of British secondary school children consume junk food daily 6 and nearly 50% of French children prefer sugar-sweetened drinks rather than water. 7 Sugar-sweetened drinks furthermore represent a significant source of dietary fructose, which has been associated with higher serum uric acid levels and increased blood pressure in particular in adolescents. 8
Behavioural factors, such as a reduction of the total amount of physical activity (PA) and an increase in the proportion of time spent in sedentary behaviours (SB), have also been associated with junk food consumption and body mass index (BMI) increase. 9,10 More specifically, TV viewing and other screen-based behaviours have been linked with an increased consumption of energy-dense and junk foods, of sugar-sweetened beverages, 10 and negatively associated with fruit and vegetable intake. 9,10
A recent review showed that specific dietary patterns are strongly related to the adolescent’s peer network, in which friends are more prone to share food-related attitudes and choices. 11
This recent evidence has shown not only an increase in unhealthy food consumption through adolescence but also a tendency for eating behaviours established in adolescence, both negative and positive ones, to persist into adulthood. 1,3 All of this highlights the need for a better understanding of the factors that can influence adolescents’ eating behaviours.
To our knowledge, these correlates of food choices have rarely been studied in a nationally representative sample of young people, using an internationally validated and standardized approach, in particular as regards Italy. The aim of this study, therefore, was to examine the influence of screen-based SB on unhealthy eating among a large representative sample of 11-, 13- and 15-year-old Italian and to assess the role of possible modifiers such as fruit and vegetable consumption, PA, number of friends, BMI and socio-economic status (SES).
Methods
The Italian study protocol and the questionnaire were approved (prot. CE-ISS 09/273) by the Ethics Committee of the Istituto Superiore di Sanità (Italian Ministry of Health).
Study population and design
Data were collected as part of the Health Behaviour in School-aged Children (HBSC) 2009–10 survey in Italy. HBSC is an international study involving 43 different countries and regions across and outside Europe. 12 To ensure consistency in survey instruments, consent procedure, data collection, processing procedures, and anonymity and confidentiality, each country adheres to the HBSC international study protocol. 12,13 A representative sample of 77 113 students aged 11, 13 and 15 years were recruited from 3555 school classes allocated throughout all Italian Regions. Cluster sampling was used, the school class being the primary sampling unit, or the school, if a sampling frame of classes was not available. 13 The study questionnaires were administered by ad hoc trained personnel, with a response rate of 96%. 14 As agreed in the international study protocol, adolescents not reporting their gender, age, or whose age was not within 6 months of the mean age of each age group, were excluded ( n = 5751, 7.4%). 12 A total of 58 928 (82.6% out of 71 362) adolescents were included in the study, respectively, 20 728 eleven-year-olds, 20 661 thirteen-year-olds and 17 539 fifteen-year-olds, of which 50.3% were girls.
Variables and measures
Eating habits
Eating habits were assessed using a short food-frequency questionnaire by asking participants how many times a week they usually consumed singularly: sweets (candy, chocolate); non-diet (sugared) soft drinks; crisps; fruits and vegetables. Response categories ranged from ‘never’ to ‘more than once a day, every day’. A previous study found the items to be valid and acceptable for use with adolescents. 15 For descriptive purposes, each variable was recoded into three categories of consumption: low consumption (once a week or less), moderate consumption (2–6 days a week), or high consumption (once a day every day or more). In order to quantify weekly unhealthy food consumption and to allow comparisons with previous HBSC published investigations, responses to the items on sweets, soft drinks and crisps consumption were recoded as follows:
‘never’ was coded 0, ‘less than once a week’ 0.5, ‘once a week’ 1, ‘2–4 days/week’ 3 (midpoint of the interval 2–3), ‘5–6 days/week’ 5.5 (midpoint of the interval 5–6), up to ‘once a day, every day’ 7, and ‘every day, more than once’ 8, 16 and then summed into a new variable representing unhealthy food consumption. The new variable was subsequently dichotomized below and above the median score value for use in multilevel logistic regression analyses.
Sedentary screen-based behaviours
Time spent in screen-based activities was assessed using two items: ‘For about how many hours a day do you usually: (i) watch television (including videos) and (ii) play PC-games or TV-games (Playstation, Xbox, Game-Cube, etc.) in your free time?’. Available response options used a 9-point scale from ‘None at all’; ‘approximately half an hour a day’; ‘approximately 1 h/a day’, and so on to ‘approximately 7 or more hours per day’. Questions were asked for weekdays and the weekend, then an SB index score was derived by summing the number of hours spent in a 7-day week. Previous studies conducted in similar age groups showed a good test–retest reliability and validity of this approach. 17 In accordance with Vereecken et al. , 15 the average number of hours of TV/PC exposure per day was grouped into four categories: 2/h or less per day, 2–4/h per day, 4–6/h and more than 6/h per day.
Adiposity
Adiposity was assessed by computing the BMI [weight/height 2 (kg/m 2 )] calculated using self-reported weight and height. Participants were then classified as overweight/obese when BMI was equal to or greater than the sex- and age-specific 75th percentile from Cole et al.18
Validation studies showed that self-reported vs. measured heights and weights represent a valid and acceptable estimate for population-based studies 19 including in Italy. 20
Physical activity
Moderate to vigorous physical activity (MVPA) was measured with a single item: ‘Over the past 7 days how many days were you physically active for a total of at least 60 min?’. The item was introduced with the definition of MVPA: ‘an activity that usually increases your heart rate and makes you get out of breath some of the time’. This question has been shown to have a good reliability and validity when compared with accelerometer data. 21 A score of 5 or more (days per week) classified respondents as meeting or not meeting the MVPA guidelines Physical Activity Guide Lines (PAGL). 22,23
Peer network
The number of close friends of the same sex was used as an indicator of the peer network. Participants were asked how many close friends they have with the following response options: none, one, two, three or more. Answers were dichotomized into less than, or three or more friends.
Socio-economic status
The SES of participants was assessed using the HBSC Family Affluence Scale (FAS) reflecting the material resources of the family. The FAS has been developed within the HBSC study to assess the participants’ SES, given the known difficulties of asking young people to accurately detail their parents occupation. 24 The FAS has been used and validated since its development and findings confirm that it is a valid indicator of young people’s material circumstances for use in cross-national surveys, 24,25
The scale consists of four items including family car ownership, whether children and adolescents have their own bedroom, number of holidays spent with their parents per year and family computer ownership. The obtained scores were recoded in a 3-point ordinal scale according to low, medium and high family affluence. 24
Statistical analysis
To account for the clustered structure of the sampling scheme, with students nested within schools, multilevel analyses were carried out using the IBM-SPSS (version 20) statistical software package. The primary sampling unit was the school class.
All analyses were performed separately by gender and age. Multilevel logistic regression models were performed to examine the relationships between unhealthy food consumption, as a dependent variable, and screen-based SB (television/computer exposures), healthy food consumptions (fruit and vegetables), peer relationship, BMI, PA and SES as independent variables. Binary logistic regression analyses were used to produce adjusted odds ratios (ORs) and 95% confidence intervals (95%CIs).
Results
Information on gender- and age-specific self-reported food consumption, screen-based SB, peer relationships, BMI, PA and SES is summarized in table 1 . Unhealthy food consumption, recoded as the sum of sweets, crisps and sugared drinks consumption ranged between a minimum score value of 0 (<1% of the whole sample) and a maximum value of 18 (1.3% of the sample) independently of age and sex, with a mean value of 7.75 (SD ± 3.8) and a median score value of 7.00. The number of adolescents with a consumption of unhealthy food above the median value increased with age in both sexes and was slightly higher in boys than in girls. Sweets and sugar drinks consumption increased between 11 and 15 years of age. For sugared drinks and crisps intake the majority of respondents reported a low consumption, once a week or less, with a slight increase in moderate consumption in all the ages and in both sexes.
Table 1.
11 years old
|
13 years old
|
15 years old
|
||||||||
---|---|---|---|---|---|---|---|---|---|---|
Variables
|
Boys N (%) | Girls N (%) | All N (%) | Boys N (%) | Girls N (%) | All N (%) | Boys N (%) | Girls N (%) | All N (%) | |
Sweets consumption | Low | 4004 (38.3) | 4003 (39.9) | 8007 (39.1) | 3131 (31.0) | 3117 (30.2) | 6248 (30.6) | 2104 (25.1) | 2306 (25.6) | 4410 (25.4) |
Moderate | 3612 (34.5) | 3359 (33.5) | 6971 (34.0) | 4184 (41.4) | 3925 (38.0) | 8109 (39.7) | 3602 (43.0) | 3620 (40.2) | 7222 (41.6) | |
High | 2850 (27.2) | 2662 (26.6) | 5512 (26.9) | 2798 (27.7) | 3289 (31.8) | 6087 (29.8) | 2675 (31.9) | 3073 (34.2) | 5748 (33.1) | |
Sugared drinks consumption | Low | 5493 (52.6) | 6367 (63.7) | 11 860 (58.0) | 4720 (46.5) | 5979 (57.9) | 10 699 (52.3) | 3248 (38.8) | 5136 (57.1) | 8384 (48.3) |
Moderate | 2902 (27.8) | 2181 (21.8) | 5083 (24.9) | 3369 (33.2) | 2715 (26.3) | 6084 (29.7) | 3213 (38.4) | 2422 (26.9) | 5635 (32.4) | |
High | 2048 (19.6) | 1444 (14.5) | 3492 (17.1) | 2052 (20.2) | 1625 (15.8) | 3677 (18.0) | 1909 (22.8) | 1445 (16.1) | 3354 (19.3) | |
Crisps consumption | Low | 6753 (69.9) | 7096 (77.0) | 13 849 (73.4) | 6864 (72.6) | 7383 (76.5) | 14 247 (74.5) | 5498 (71.0) | 6451 (76.8) | 11 949 (74.0) |
Moderate | 1992 (20.6) | 1532 (16.6) | 3524 (18.7) | 2008 (21.2) | 1799 (18.6) | 3807 (19.9) | 1819 (23.5) | 1609 (19.2) | 3428 (21.2) | |
High | 921 (9.5) | 587 (6.4) | 1508 (8.0) | 589 (6.2) | 475 (4.9) | 1064 (5.6) | 428 (5.5) | 340 (4.1) | 768 (4.8) | |
Unhealthy food a consumption (sweets, coke, crisps) | Below the median | 5832 (61.5) | 6233 (68.8) | 12 065 (65.1) | 5487 (58.9) | 6053 (63.6) | 11 540 (61.3) | 4092 (53.6) | 5202 (62.9) | 9294 (58.5) |
Over the median | 3647 (38.5) | 2821 (31.2) | 6468 (34.9) | 3823 (41.1) | 3460 (36.4) | 7283 (38.7) | 3536 (46.4) | 3066 (37.1) | 6602 (41.5) | |
Screen-basedSB | ≤2 h a day | 4139 (40.5) | 4423 (44.8) | 8562 (42.6) | 2411 (24.0) | 2112 (20.6) | 4523 (22.3) | 1597 (19.2) | 1529 (17.0) | 3126 (18.1) |
To 4 h a day | 3546 (34.7) | 3375 (34.2) | 6921 (34.5) | 3841 (38.3) | 3814 (37.1) | 7655 (37.7) | 3239 (38.9) | 3356 (37.4) | 6595 (38.1) | |
To 6 h a day | 1494 (14.6) | 1306 (13.2) | 2800 (13.9) | 2239 (22.3) | 2510 (24.4) | 4749 (23.4) | 2029 (24.4) | 2278 (25.4) | 4307 (24.9) | |
>6 h a day | 1040 (10.2) | 767 (7.8) | 1807 (9.0) | 1540 (15.4) | 1842 (17.9) | 3382 (16.7) | 1458 (17.5) | 1814 (20.2) | 3272 (18.9) | |
Vegetables consumption | Low | 4215 (40.4) | 3200 (31.9) | 7415 (36.2) | 3576 (35.2) | 3096 (29.9) | 6672 (32.5) | 2778 (33.1) | 2500 (27.7) | 5278 (30.3) |
Moderate | 3909 (37.4) | 3992 (39.8) | 7901 (38.6) | 4414 (43.5) | 4468 (43.1) | 8882 (43.3) | 3913 (46.6) | 3991 (44.2) | 7904 (45.4) | |
High | 2322 (22.2) | 2827 (28.2) | 5149 (25.2) | 2164 (21.3) | 2795 (27.0) | 4959 (24.2) | 1708 (20.3) | 2538 (28.1) | 4246 (24.4) | |
Fruits consumption | Low | 2318 (22.1) | 1928 (19.2) | 4246 (20.7) | 2087 (20.5) | 2147 (20.7) | 4234 (20.6) | 1890 (22.5) | 2029 (22.4) | 3919 (22.4) |
Moderate | 3635 (34.6) | 3232 (32.2) | 6867 (33.4) | 4093 (40.2) | 3596 (34.7) | 7689 (37.4) | 3487 (41.5) | 3109 (34.3) | 6596 (37.8) | |
High | 4544 (43.3) | 4889 (48.7) | 9433 (45.9) | 4006 (39.3) | 4631 (44.6) | 8637 (42.0) | 3034 (36.1) | 3918 (43.3) | 6952 (39.8) | |
PA | <1 h/day | 7943 (80.1) | 8204 (87.2) | 16 147 (83.5) | 7775 (80.2) | 8404 (88.1) | 16 179 (84.1) | 6396 (81.4) | 6893 (88.7) | 13 289 (85.0) |
1 h every day | 1978 (19.9) | 1208 (12.8) | 3186 (16.5) | 1926 (19.9) | 1141 (12.0) | 3067 (15.9) | 1465 (18.6) | 877 (11.3) | 2342 (15.0) | |
Close friends of the same sex | <3 | 1626 (15.5) | 4485 (45.4) | 6111 (30.0) | 2149 (21.1) | 5202 (50.6) | 7351 (35.9) | 2135 (25.4) | 5646 (62.8) | 7781 (44.7) |
3 or more | 8863 (84.5) | 5404 (54.7) | 14 267 (70.0) | 8046 (78.9) | 5078 (49.4) | 13 124 (64.1) | 6283 (74.6) | 3342 (37.2) | 9625 (55.3) | |
BMI | Normal weight | 6173 (76.8) | 6018 (83.7) | 12 191 (80.0) | 6040 (78.6) | 7611 (86.3) | 13 651 (82.7) | 5905 (77.9) | 7367 (89.2) | 13 272 (83.8) |
Overweight/obese | 1868 (23.2) | 1175 (16.3) | 3043 (20.0) | 1646 (21.4) | 1212 (13.7) | 2858 (17.3) | 1672 (22.1) | 896 (10.8) | 2568 (16.2) | |
SES | FAS low | 1318 (12.9) | 1605 (16.2) | 2923 (14.6) | 953 (9.5) | 1258 (12.3) | 2211 (10.9) | 778 (9.4) | 1019 (11.4) | 1797 (10.4) |
FAS middle | 4604 (45.1) | 4570 (46.3) | 9174 (45.7) | 4045 (40.2) | 4382 (42.7) | 8427 (41.5) | 3271 (39.3) | 3737 (41.8) | 7008 (40.6) | |
FAS high | 4287 (42.0) | 3706 (37.5) | 7993 (39.8) | 5060 (50.3) | 4632 (45.1) | 9692 (47.7) | 4266 (51.3) | 4190 (46.8) | 8456 (49.0) |
Note : Consumptions: Low, once a week or less; Moderate, 2–6 times a week; High, daily or more.
BMI: below and above the sex-, and age-specific 75th percentile from Cole et al.18
a Sweets, sugared drinks and crisps summed and dichotomized below and over the median values.
Adolescents spent, on average, 3–4 h a day in screen-based behaviours [mean hours per day for boys vs. girls were, respectively: 3.07 vs. 2.83 in 11 y.o.; 3.82 vs. 4.07 in 13 y.o. and 4.08 vs. 4.26in 15y.o. (data not shown)] and girls were the most sedentary among the 13- and 15-year-olds. On average, 57% of 11-year-olds spent more than 2 h a day in screen-based behaviours (watching TV or using a PC), and this increased with age to more than 80% of 15-year-olds.
Vegetable consumption was the lowest among boys and in the youngest while fruit consumption was higher among 11-year-olds and girls.
The percentage of adolescents meeting PAGL was lower than 20%, and further declined with age. Girls were the least active, in the oldest group 89% of girls did not meet the PA recommendations.
The proportion of adolescents reporting three or more close friends of the same sex decreased with age, falling from 70% in 11-year-olds to 55% of 15-year-olds. Girls reported having fewer close friends than boys.
BMI was higher in among boys across all ages, but particularly for 11-year-olds. As regards SES, the large majority of the sample reported medium or high family affluence.
The logistic regression analyses results predicting the risk of being in the unhealthy eating group (above the population median) are reported in table 2 .
Table 2.
Boys 11yo | Girls 11yo | All 11yo | Boys 13yo | Girls 13yo | All 13yo | Boys 15yo | Girls 15yo | All 15yo | ||
---|---|---|---|---|---|---|---|---|---|---|
OR (95%CI) | OR (95%CI) | OR (95%CI) | OR (95%CI) | OR (95%CI) | OR (95%CI) | OR (95%CI) | OR (95%CI) | OR (95%CI) | ||
Screen-based SB | ≤2 h a day | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
To 4 h a day | 1.56 (1.24;1.96) | 2.09 (1.65;2.65) | 1.77 (1.52;2.06) | 1.65 (1.25;2.17) | 1.38 (1.04;1.82) | 1.51 (1.25;1.83) | 1.12 (0.85;1.49) | 2.39 (1.71;3.34) | 1.57 (1.27;1.95) | |
To 6 h a day | 2.07 (1.57;2.71) | 2.48 (1.86;3.31) | 2.23 (1.80;2.75) | 2.67 (1.99;3.59) | 2.20 (1.65;2.93) | 2.43 (1.99;2.96) | 1.54 (1.14;2.09) | 2.90 (2.06;4.07) | 2.00 (1.60;2.50) | |
>6 h a day | 3.80 (2.69;5.36) | 5.25 (3.44;8.01) | 4.31 (3.29;5.65) | 3.44 (2.53;4.67) | 4.32 (3.15;5.94) | 3.89 (3.14;4.82) | 2.49 (1.81;3.41) | 5.10 (3.61;7.21) | 3.44 (2.73;4.33) | |
Vegetables consumption | Low | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Moderate | 0.61 (0.50;0.75) | 0.71 (0.56;0.91) | 0.66 (0.56;0.77) | 0.67 (0.54;0.82) | 0.68 (0.53;0.87) | 0.68 (0.58;0.79) | 0.80 (0.65;0.98) | 0.60 (0.49;0.73) | 0.70 (0.61;0.81) | |
High | 0.45 (0.33;0.62) | 0.56 (0.42;0.76) | 0.50 (0.40;0.63) | 0.51 (0.38;0.68) | 0.46 (0.34;0.61) | 0.49 (0.40;0.60) | 0.69 (0.51;0.94) | 0.42 (0.32;0.55) | 0.54 (0.44;0.66) | |
Fruits consumption | Low | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Moderate | 0.98 (0.75;1.28) | 0.72 (0.54;0.96) | 0.87 (0.71;1.06) | 0.62 (0.48;0.82) | 1.08 (0.82;1.41) | 0.82 (0.67;0.99) | 0.97 (0.76;1.24) | 0.98 (0.76;1.27) | 0.98 (0.82;1.16) | |
High | 0.91 (0.71;1.18) | 0.86 (0.65;1.13) | 0.88 (0.73;1.07) | 0.68 (0.51;0.90) | 0.88 (0.68;1.13) | 0.77 (0.63;0.94) | 0.73 (0.56;0.95) | 0.70 (0.53;0.92) | 0.71 (0.59;0.86) | |
PA | <1 h/day | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
1 h every day | 1.37 (1.07;1.76) | 0.85 (0.63;1.15) | 1.17 (0.96;1.43) | 1.26 (0.97;1.64) | 0.97 (0.70;1.33) | 1.13 (0.93;1.37) | 0.98 (0.78;1.22) | 1.12 (0.81;1.53) | 1.04 (0.86;1.24) | |
Close friendsof the same sex | <3 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
3 or more | 1.61 (1.24;2.09) | 1.04 (0.80;1.35) | 1.27 (1.05;1.54) | 1.34 (1.04;1.73) | 1.05 (0.85;1.28) | 1.17 (0.99;1.38) | 1.12 (0.92;1.36) | 1.23 (1.01;1.49) | 1.17 (1.02;1.35) | |
BMI | Normal weight | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Overweight/obese | 0.67 (0.56;0.81) | 0.95 (0.74;1.22) | 0.76 (0.65;0.88) | 0.64 (0.49;0.82) | 0.59 (0.44;0.79) | 0.62 (0.52;0.74) | 0.56 (0.45;0.69) | 0.60 (0.42;0.84) | 0.57 (0.47;0.69) | |
SES | FAS low | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
FAS middle | 0.89 (0.66;1.21) | 0.66 (0.51;0.87) | 0.77 (0.63;0.94) | 1.06 (0.73;1.53) | 1.09 (0.81;1.46) | 1.07 (0.85;1.35) | 0.80 (0.57;1.12) | 0.74 (0.52;1.04) | 0.76 (0.59;0.98) | |
FAS high | 0.95 (0.70;1.31) | 0.79 (0.60;1.04) | 0.86 (0.70;1.07) | 0.89 (0.63;1.27) | 0.99 (0.74;1.34) | 0.94 (0.74;1.18) | 0.83 (0.60;1.14) | 0.66 (0.46;0.96) | 0.73 (0.57;0.94) |
Notes: Consumptions: Low, once a week or less; Moderate, 2–6 times a week; High, daily or more.
BMI: below and over the sex- and age-specific 75th percentile from Cole et al.18 Bold entries are reported ORs with significant 95%CI.
a Sweets, beverage and crisps summed and dichotomized below and over the median value.
Screen-based SB were strongly associated with unhealthy food consumption. The risk of consuming junk food increased with increasing time spent watching TV or using a PC, shifting in 11y.o. children for example, from an OR of 1.77 (95%CI 1.52;2.06) for 2–4 h/day to an OR of 4.31 (95%CI 3.29;5.65) for more than 6 h/day. A similar pattern was found for 13- and 15-year-olds. Girls showed the highest risk in all age groups.
A lower intake of fruit and, in particular, of vegetables was significantly associated with the likelihood of being in the unhealthy eating group. In all the age groups and in both sexes, the odds of unhealthy food consumption decreased as fruit and vegetable intake increased.
Junk food consumption was not associated with PA, except among 11-year-old boys (OR = 1.37, 95%CI 1.07; 1.76).
Having three or more close friends of the same gender was also associated with junk food consumption in both sexes: the highest risk was observed in 11- and 13-year-old boys and in 15-year-old girls.
BMI was found to be inversely associated with unhealthy food consumption. This association was observed in all ages and both sexes, except for 11-year-old girls. No consistent relationship was found between SES and consumption of unhealthy food.
Discussion
In the present study, correlates of unhealthy eating behaviour were investigated in a representative sample of Italian 11-, 13- and 15-year-old young adolescents.
Previously, time spent in SB, and TV viewing in particular, has been strongly linked with negative eating patterns. 26,27 In accordance with these studies, our results show a clear positive association of unhealthy food consumption and time spent in screen-based behaviours, and a negative association of unhealthy food consumption and the consumption of healthy foods, particularly vegetables.
During adolescence, parents provide the main contextual environment for many behaviours and in particular television viewing, computer playing and food consumption within the home. If children are left in front of the screen with available foods and drinks to distract them, they may associate television viewing with food intake. Indeed, when TV viewing is repeatedly accompanied by consumption of snacks and sugar-sweetened beverages, the screen-viewing itself may thus become an automatic cue to junk food intake also in the absence of hunger. 27 Furthermore, children’s food preferences, requests and consumption have been shown to be influenced by marketing proposals. During time spent watching TV, adolescents are exposed to varied food advertisements that can proportionally influence their choices. 28 In fact, studies have found that each additional hour of television watched can result in an increased servings of junk foods, commonly advertised on television and a reduction of almost one fruit and vegetable serving per week. 27,29,30
Adolescents are also increasingly exposed to a wide range of influences beyond the family environment including through school, the local community, media and social media. Children and adolescents alike have also been shown to be influenced by what their peers eat. 31 In accordance with these findings, our results showed an increased risk of eating unhealthy food among those with a wide peer network, particularly younger boys.
Our findings also revealed an interesting association between unhealthy food consumption and BMI. Recent studies that explored the relationship between BMI and eating patterns have reported inconsistent results, the large majority of these showing no association between increased BMI and junk food or sweetened drinks consumption. 32 Only a few studies have shown an increase in junk food intake among overweight children. 33 In contrast, we found that the likelihood of being a heavy consumer of unhealthy foods was significantly higher among normal weight adolescents. One explanation for this finding maybe the different control exerted by the parents of overweight adolescents compared with those of normal weight. It is therefore essential that nutritional programmes are targeted at all adolescents and not just those who are currently overweight or obese.
Recently there has been an increased interest in parents feeding practices. 34 In particular, mothers of overweight or obese children seem to exert an active role in reducing the likelihood of consuming unhealthy fatty food. 35 Above all, and independent of child BMI, the use of feeding strategies seems to be dependent on parents’ educational and economic background such that the higher the familial SES, the higher the exerted control. 36 Our results were also suggestive of a lower risk of being a heavy unhealthy food consumer in the highest SES, but with inconsistent statistical significance. 37
Strengths and limitations
The cross-sectional nature of the HBSC study means that we are unable to make any inferences about causal relationships, in particular in identifying specific determinants of eating behaviours. In addition, the study was limited to participants in three specific ages. In relation to screen-based SB, we used proxy measures of time spent in watching TV and using a computer for leisure as two distinct behaviours. Recent evidence, however, suggests that an increased proportion of youth engage in multitasking SB, which means adolescents could differently declare behaviours that happens simultaneously (i.e. surfing Internet or chatting online while watching TV) with further differences among genders. 38 More work is needed to better understand the health impact of engaging in simultaneous sedentary activities.
Strengths of this study were the use of a standardized approach and the representativeness of the sample, increasing both the study validity and its generalizability. No other published research has examined these predictive correlates of behaviours in such a large national representative samples of adolescents in Italy.
Conclusions
In conclusion, our results highlight the importance of recognizing the complexity of behaviours linked to food consumption in adolescence in order to drive and inform further effective nutrition and public health interventions. This means considering diet and nutrition, independently of adiposity status, as an objective of nutritional public health initiatives, especially among those who engage in excessive screen-based SB. These findings suggest that nutritional advice should be aimed both at decreasing the consumption of soft drinks, fast food and snack food and at addressing the role of television in sending opposite messages. 28 Since school and parents provide the child’s main contextual environment, 39 they can be viewed as key players and central agents of change in the prevention of unhealthy eating. More work is needed to explore how school and parents can effectively support adolescents in developing their ability to access, analyse, evaluate, and appropriately use media and technologies to which they are exposed. Parents should therefore be involved in tailored educational initiatives aimed at increasing their awareness of the impact unhealthy food can have on adolescents’ health, the influence that television viewing can have on their familial food choices, and of the importance of proactively contributing to the way their children manage their screen time behaviour. Public health efforts should be oriented towards effective health policies for the regulation of television advertising aimed at children.
Acknowledgements
HBSC is a World Health Organization cross-national study. The international coordinator for the 2009/2010 study was Candace Currie, University of St Andrews; Data Bank Manager: Oddrun Samdal, University of Bergen. In Italy the study has been carried out since 2001 under the coordination of the Universities of Torino, Padova and Siena. In the 2009/2010 survey the study has been conducted with the supervision of the National Institutes of Health and the Ministry of Health, under the lead of Professor F. Cavallo (Principal Investigator for Italy).
Funding
This work is part of the Project ‘ Sistema di indagini sui rischi comportamentali in età 6-17 anni ’ promoted and financed by the Ministry of Health (cap. 4393/2005-CCM).
Conflicts of interest : None declared.
Increased time spent in screen-based behaviours in adolescence significantly increases the consumption of unhealthy food.
The increase in unhealthy food consumption decreases the likelihood of consuming healthy food in particular vegetables.
The likelihood of being a heavy consumer of unhealthy food is higher among normal weight than in overweight adolescents.
Diet and nutritional interventions targeting adolescents should address both eating behaviours and screen-based behaviours independently of adiposity status.
Public health efforts should be oriented towards effective health policies for the regulation of television advertising aimed at children and adolescents should actively support actions to develop youth’s ability to access and appropriately use media and technologies to which they are exposed.
References
- 1. Bremer A, Lustig R . Effects of sugar-sweetened beverages on children . Pediatr Ann 2012. ; 41 : 26 – 30 . [DOI] [PubMed] [Google Scholar]
- 2. Ludwig DS, Peterson KE, Gortmaker SL . Relation between consumption of sugar-sweetened drinks and childhood obesity: a prospective, observational analysis . The Lancet 2001. ; 357 : 505 – 8 . [DOI] [PubMed] [Google Scholar]
- 3. Litin L, Sacks F . Trans-fatty-acid content of common foods . N Engl J Med 1993. ; 329 : 1969 – 70 . [DOI] [PubMed] [Google Scholar]
- 4. Mensink RP, Katan MB . Effect of dietary trans fatty acids on high-density and low-density lipoprotein cholesterol levels in healthy subjects . N Engl J Med 1990. ; 323 : 439 – 45 . [DOI] [PubMed] [Google Scholar]
- 5. Nestle M . Food marketing and childhood obesity—a matter of policy . N Engl J Med 2006. ; 354 : 2527 – 9 . [DOI] [PubMed] [Google Scholar]
- 6. Zahra J, Ford T, Jodrell D . Cross‐sectional survey of daily junk food consumption, irregular eating, mental and physical health and parenting style of British secondary school children . Child Care Health Dev 2014. ; 40 : 481 – 91 . [DOI] [PubMed] [Google Scholar]
- 7. Isacco L, Lazaar N, Ratel S , et al. . The impact of eating habits on anthropometric characteristics in French primary school children . Child Care Health Dev 2010. ; 36 : 835 – 42 . [DOI] [PubMed] [Google Scholar]
- 8. Nguyen S, Choi HK, Lustig RH, Hsu C-y . Sugar-sweetened beverages, serum uric acid, and blood pressure in adolescents . J Pediatr 2009. ; 154 : 807 – 13 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Barr-Anderson DJ, Larson NI, Nelson MC , et al. . Does television viewing predict dietary intake five years later in high school students and young adults? Int J Behav Nutr Phys Act 2009. ; 6 : 8 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Tremblay MS, LeBlanc AG, Kho ME , et al. . Systematic review of sedentary behaviour and health indicators in school-aged children and youth . Int J Behav Nutr Phys Act 2011. ; 8 : 98 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Haye K, Robins G, Mohr P, Wilson C . Adolescents’ intake of junk food: processes and mechanisms driving consumption similarities among friends . J Res Adolesc 2013. ; 23 : 524 – 36 . [Google Scholar]
- 12. Currie C, Nic Gabhainn S, Godeau E . The health behaviour in school-aged children: WHO Collaborative Cross-National (HBSC) study: origins, concept, history and development 1982–2008 . Int J Public Health 2009. ; 54 : 131 – 9 . [DOI] [PubMed] [Google Scholar]
- 13. Roberts C, Freeman J, Samdal O , et al. . The Health Behaviour in School-aged Children (HBSC) study: methodological developments and current tensions . Int J Public Health 2009. ; 54 : 140 – 50 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Lazzeri G, Giacchi M, Dalmasso P , et al. . The methodology of the Italian HBSC 2010 study (Health Behaviour in School-aged Children) . Ann Ig 2013. ; 25 : 225 – 33 . [DOI] [PubMed] [Google Scholar]
- 15. Vereecken C, Todd J, Roberts C , et al. . Television viewing behaviour and associations with food habits in different countries . Public Health Nutr 2006. ; 9 : 244 – 50 . [DOI] [PubMed] [Google Scholar]
- 16. Vereecken CA, Rossi S, Giacchi MV, Maes L . Comparison of a short food-frequency questionnaire and derived indices with a seven-day diet record in Belgian and Italian children . Int J Public Health 2008. ; 53 : 297 – 305 . [DOI] [PubMed] [Google Scholar]
- 17. Schmitz KH, Harnack L, Fulton JE , et al. . Reliability and validity of a brief questionnaire to assess television viewing and computer use by middle school children . J Sch Health 2004. ; 74 : 370 – 7 . [DOI] [PubMed] [Google Scholar]
- 18. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH . Establishing a standard definition for child overweight and obesity worldwide: international survey . BMJ 2000. ; 320 : 1240 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Spencer EA, Appleby PN, Davey GK, Key TJ . Validity of self-reported height and weight in 4808 EPIC–Oxford participants . Public Health Nutr 2002. ; 5 : 561 – 5 . [DOI] [PubMed] [Google Scholar]
- 20. Dalmasso P, Charrier L, Zambon A , et al. . Does self reported BMI really reflect the proportion of overweight and obese children? Biomed Stat Clin Epidemiol 2010. ; 4 : 7 – 13 . [Google Scholar]
- 21. Booth M, Okely A, Chey T, Bauman A . The reliability and validity of the physical activity questions in the WHO health behaviour in schoolchildren (HBSC) survey: a population study . Br J Sports Med 2001. ; 35 : 263 – 7 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Corbin CB, Pangrazi RP, Le Masurier GC . Physical activity for children: Current patterns and guidelines . J Phys Act Health 2004. ; 1 : 281 . [Google Scholar]
- 23. Borraccino A, Lemma P, Iannotti R , et al. . Socio-economic effects on meeting PA guidelines: comparisons among 32 countries . Med Sci Sports Exerc 2009. ; 41 : 749 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Currie C, Molcho M, Boyce W , et al. . Researching health inequalities in adolescents: the development of the Health Behaviour in School-Aged Children (HBSC) family affluence scale . Soc Sci Med 2008. ; 66 : 1429 – 36 . [DOI] [PubMed] [Google Scholar]
- 25. Currie CE, Elton RA, Todd J, Platt S . Indicators of socioeconomic status for adolescents: the WHO Health Behaviour in School-aged Children Survey . Health Educ Res 1997. ; 12 : 385 – 97 . [DOI] [PubMed] [Google Scholar]
- 26. Wiecha JL, Peterson KE, Ludwig DS , et al. . When children eat what they watch: impact of television viewing on dietary intake in youth . Arch Pediatr Adolesc Med 2006. ; 160 : 436 – 42 . [DOI] [PubMed] [Google Scholar]
- 27. Kremers SP, van der Horst K, Brug J . Adolescent screen-viewing behaviour is associated with consumption of sugar-sweetened beverages: the role of habit strength and perceived parental norms . Appetite 2007. ; 48 : 345 – 50 . [DOI] [PubMed] [Google Scholar]
- 28. Chandon P, Wansink B . Does food marketing need to make us fat? A review and solutions . Nutr Rev 2012. ; 70 : 571 – 93 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Harris JL, Bargh JA, Brownell KD . Priming effects of television food advertising on eating behavior . Health Psychol 2009. ; 28 : 404 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Klepp K-I, Wind M, De Bourdeaudhuij I , et al. . Television viewing and exposure to food-related commercials among European school children, associations with fruit and vegetable intake: a cross sectional study . Int J Behav Nutr Phys Act 2007. ; 4 : 46 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Patrick H, Nicklas TA . A review of family and social determinants of children’s eating patterns and diet quality . J Am Coll Nutr 2005. ; 24 : 83 – 92 . [DOI] [PubMed] [Google Scholar]
- 32. Yannakoulia M, Brussee SE, Drichoutis AC , et al. . Food consumption patterns in Mediterranean adolescents: are there differences between overweight and normal-weight adolescents? J Nutr Educ Behav 2012. ; 44 : 233 – 9 . [DOI] [PubMed] [Google Scholar]
- 33. Lioret S, Touvier M, Lafay L , et al. . Dietary and physical activity patterns in French children are related to overweight and socioeconomic status . J Nutr 2008. ; 138 : 101 – 7 . [DOI] [PubMed] [Google Scholar]
- 34. McCurdy K, Gorman KS . Measuring family food environments in diverse families with young children . Appetite 2010. ; 54 : 615 – 8 . [DOI] [PubMed] [Google Scholar]
- 35. Crouch P, O'DEA JA, Battisti R . Child feeding practices and perceptions of childhood overweight and childhood obesity risk among mothers of preschool children . Nutr Diet 2007. ; 64 : 151 – 8 . [Google Scholar]
- 36. Kröller K, Warschburger P . Associations between maternal feeding style and food intake of children with a higher risk for overweight . Appetite 2008. ; 51 : 166 – 72 . [DOI] [PubMed] [Google Scholar]
- 37. McPhie S, Skouteris H, Daniels L, Jansen E . Maternal correlates of maternal child feeding practices: a systematic review. Matern Child Nutr 2014. ; 10 : 18 – 43 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Foehr UG . Media multitasking among American youth: prevalence, predictors and pairings. Menlo Park, CA: Henry J Kaiser Family Foundation 2006. .
- 39. Lemma P, Borraccino A, Berchialla P , et al. . Well-being in 15-year-old adolescents: a matter of relationship with school . J Public Health 2015. ; 37 : 573 – 80 . [DOI] [PMC free article] [PubMed] [Google Scholar]