Promoting health behaviour in Portuguese children via Short Message Service: The efficacy of a text-messaging programme

Abstract

A Short Message Service programme was adapted to monitor three health behaviours and provide supportive feedback. The study aimed to evaluate the effectiveness of the programme to increase fruit/vegetable consumption and physical activity and to decrease screen time. A total of 139 Portuguese children, aged 8–10 years, grouped by classroom, were randomly assigned to an intervention (8 weeks of monitoring/feedback) or a control condition. Participants had their key behaviours assessed at baseline, post-intervention and follow-up. A three-level hierarchical linear model was developed. Results showed that the monitoring and feedback programme significantly increased fruit and vegetable consumption over time.

Keywords

children health promotion healthy behaviours technology

Introduction

In most countries, the prevalence of obesity and overweight has doubled in the last decades (Lobstein et al., 2010; Seidell, 2005). In Portugal, the prevalence of obesity and overweight has increased from 49.6 per cent (in 1995–1998) to 53.6 per cent (in 2003–2005) (Do Carmo et al., 2008). And, a recent study (Sardinha et al., 2010) using World Health Organization (WHO) cut-offs revealed that 23.5 per cent of the boys aged from 10 to 18 years were overweight (5.8% obese) and 21.6 per cent of girls were overweight (4.6% obese). This increase in obesity prevalence, especially in cities (Lobstein et al., 2010), has become a serious public health concern (Seidell, 2005), due to associated physical (e.g. diabetes), psychological (e.g. low self-esteem) and social (e.g. social isolation) problems.

The WHO declared that healthy diets and regular, adequate physical activity are major factors in the promotion and maintenance of good health throughout the entire life span. The WHO recommends that everyone should engage in at least 30 minutes of moderate physical activity every day. Fruit and vegetables are important components of a healthy diet, and their daily consumption can help to prevent major diseases. Obesity and overweight prevention programmes should then focus on the promotion of health behaviours and target children and adolescents (Uauy et al., 2010).

Several studies have shown that technology-enhanced measures could be a useful and innovative tool in promoting health behaviours. For example, text-messaging can be used as an assessment tool (Carter et al., 2012) to evaluate behaviours or states without any additional instruments. Some advantages of these tools are their accessibility at any time, fast and instant information, and ability to transmit guidance and advice without major effort and low costs (Bauer et al., 2010). Some limitations include costs of device ownership, possible margin-alization of some participants (e.g. illiterate participants or people with no financial possibilities) and, also, the fact that the intervention can be interrupted very easily for technical reasons (Cole-Lewis and Kershaw, 2010). Despite these limitations, text-messaging is still a communication tool globally used, that has showed promising results in promoting health behaviours.

Bauer et al. (2003) developed an Internet-based computer programme that uses text-messaging as a tool to provide support and feedback based on specific goals. And, most recently, Shapiro et al. (2008) showed that this text-messaging programme might be a useful tool for self-monitoring sugar-sweetened beverages, physical activity and screen time in children aged from 5 to 13 years.

This programme includes methods from social cognitive theory and behavioural models. For instance, self-monitoring and immediate feedback based on specific goals are important elements in behavioural theory that were used in the programme to promote health behaviour change. Our research team adapted the same programme to promote three health behaviours (fruit and vegetable consumption, physical activity and screen time) in children aged from 8 to 10 years. Based on the reported daily input of participants, supportive feedback messages were automatically sent via the programme. The aim of this minimal intervention was to promote behavioural change. Specifically, it intended to increase consumption of fruits and vegetables and the number of steps per day (physical activity) and to decrease screen time.

This study aimed to assess the effectiveness of the Short Message Service (SMS) programme to promote three health behaviours among Portuguese children. The primary question addressed was whether the intervention group (SMS monitoring), in comparison to the control group (no monitoring or feedback), improves the frequency of behaviours over time.

Methods

Participants

Recruitment took place at four elementary public schools from the same school district. All schools were located in the city centre of Braga, in the same geographical urban area. No criteria were defined for school’s selection. Eight elementary school classrooms with a total of 139 children (66 girls and 73 boys, aged 8–10 years) were randomly assigned to an intervention (8 weeks of SMS monitoring and supportive feedback) (n = 69), or to a control condition (n = 70) by tossing of a coin. The mean age of the sample was 9.9 (standard deviation (SD) = 0.4).

Measures

Food frequency and physical activity questionnaire

A self-report measure was used to assess eating habits (especially, fruit and vegetable consumption) and the frequency of physical activity and screen time of participants. The intake of fruits and vegetables was measured using questions of a Food-Frequency Questionnaire (FFQ) that showed good reliability and validity (Vereecken and Maes, 2003) and was based on the ‘Health Behaviour in School-Aged Children Study’ (HBSC) cross-national study (Currie et al., 2008). Vereecken and Maes (2003) showed that the test and retest reliability ranged from 0.43 to 0.70. Results showed good agreement between the FFQ and a 24-hour Food Behaviour Checklist (FBC) for most items (Vereecken and Maes, 2003). In this study, children were asked ‘How many times a week do you usually eat/drink …’ followed by a list of food and beverage items, including ‘fruits’ and ‘vegetables’. They could select, in a 5-point Likert scale, the options ‘never’, ‘less than once a week’, ‘once a week’, ‘2–4 days a week’, ‘5–6 days a week’, ‘once every day’ and ‘several times every day’ as answers. In addition, one question, ‘How many time a day do you eat fruit/vegetables’, was added to this version in a 5-point Likert scale from ‘0’ to ‘5 or more’. Daily physical activity and screen time were measured using two questions from the Family Eating and Activity Habits Questionnaire (FEAHQ) (Golan and Weizman, 1998): ‘How many hours per day on average did you participate in activities such as fast walking, swimming, ball games, etc.?’; ‘How many minutes did you spend in front of the screen (e.g. TV, computer, video games etc.)?’. The FEAHQ is a reliable questionnaire (test and retest reliability scores for individual items and for the total score ranged from 0.78 to 0.90) and internally consistent (Mean r was 0.83) (Golan and Weizman, 1998). Responses to the activity scale were given in number of hours and these were added for each item (Golan and Weizman, 1998).

Pedometers

Each participant from the intervention group received a pedometer (Silva Pedometer Plus Art. no. 56913) to count the number of daily steps.

Body mass index

Weight and height of each participant were measured in the classroom by the researcher at recruitment using a digital scale and a stadiometer. That information was subsequently used to calculate the body mass index (BMI) (kg/m²) for age. For this study, we used standard deviation scores (SDS), as recommended by Cole (2002).

Self-report satisfaction questionnaire

Children from the intervention group completed a satisfaction questionnaire with eight questions (e.g. ‘How much fun did you have doing this program’) after the intervention in order to explore children’s overall satisfaction with specific components of the programme (e.g. appropriateness of feedback messages). The response options were, for example, ‘no fun at all’, ‘no fun’, ‘neither nor’, ‘fun’, ‘lot of fun’.

Procedure and study design

Informed consent to participate in the study and to use own private cell phone was obtained from parents. All parents and/or their child, from the intervention group, had access to at least one cell phone that was registered in the study. Subsequently, after the informed consent, school classes were randomly assigned to control or to intervention condition. After the first assessment (baseline – T1), participants from both conditions were evaluated at two additional assessment points: 8 weeks later (after intervention period – T2) and, again, 4 weeks after the second evaluation (follow-up – T3). At baseline, all children attended two 60-minute psycho-educational sessions presented in a group format by two trained psychologists, while parents participated in one educational session (before the children’s participation). The sessions provided information and instruction on the three targeted behaviours (fruit and vegetable consumption, physical activity and screen time) for all participants. Session 1 focused on the benefits of increasing physical activity, decreasing screen time and the risks of sedentary behaviour. Session 2 focused on healthy diet in general and the importance of including fruits and vegetables in daily meals. The only difference between conditions was that, in session 2, children in the intervention condition received information about the SMS programme.

Children in the intervention condition monitored their behaviour for 8 weeks providing information (number of steps achieved, measured with a pedometer; number of fruits and vegetables consumed; and screen time in minutes) via SMS once at the end of the day. The specific questions were (1) How many fruits and vegetables did you eat today? (2) How many steps did you achieve with the pedometer today? and (3) How many minutes did you spend in front of a screen today? Previous research or guidelines were used to define the specific behavioural goals: five portions of fruits and vegetables (WHO, 2003), 10,000 steps per day (Tudor-Locke et al., 2004, 2008) and less than 60 minutes of screen time per day.

Incoming messages were sent to a modem that was connected to a secure web server. The messages were then evaluated according to an algorithm based on how many goals were met and improvement or deterioration in behaviour from the previous day. Automatically, the programme suggested tailored feedback messages that were sent to each child. A pool of 900 existing feedback messages from previous research were translated from English to Portuguese and adapted into the context of fruit and vegetable intake (Shapiro et al., 2008). These messages aimed to motivate the children to reach all the behaviour goals and support and reinforce positive development based on improvement or deterioration compared to the previous day (e.g. ‘Great, you met your goal for physical activity and screen time! What happened to fruits and vegetables?’). The SMS system did not allow comments or replies to the feedback messages. Details of the programme and the algorithm used have been described elsewhere (Bauer et al., 2003; Shapiro et al., 2008).

All costs incurred in mobile phone usage programme were reimbursed; however, no further incentives were provided. The Ethics Committee of the School of Psychology approved the study protocol and all parents/legal guardians signed authorization and informed consent.

Analyses

Initially, chi-square and independent samples t-tests were used to compare participants from the intervention and control condition, at baseline, and on gender, age and BMI. To assess the efficacy of the text-messaging programme to promote health behaviours (i.e. increase fruit and vegetable intake and the number of steps and to decrease screen time), we developed a three-level hierarchical linear model (HLM) analysis using SPSS (version 22), with repeated observations (Level 1) nested within participants (Level 2), which were further nested within classrooms (Level 3). Estimate parameters for all models were considered to analyse the growth over time between groups. Values of p < .05 were considered statistically significant.

The HLM analysis provides the following tests of interest: (1) The main effect for Group provides a comparison between the intervention and control groups at baseline. (2) The main effects for Time and Time-Squared provide tests of whether the control group changed significantly from T1 to T2 and from T2 to T3, respectively. (3) The Group-by-Time and Group-by-Time-Squared interactions provide tests of whether the change from T1 to T2 and from T2 to T3, respectively, differed significantly between intervention and control groups.

Results

Table 1 presents demographic variables of the sample. No significant differences between the intervention and control conditions were found for gender, χ² (1, N = 139) = .574, p = .449, and age, t = .871, degrees of freedom (df) = 137, p = .385. According to the International Obesity Task Force (IOTF) cut-offs, 39.6 per cent of the total sample was overweight or obese. No significant differences regarding the BMI-SDS were found between the two groups, t = −.747, df = 134, p = .45.

Table 1.

Demographic and baseline characteristics for each condition and the total number of participants.

	IG (N = 69)		CG (N = 70)		All (N = 136)
Gender^a	♀ 35	(50.7)	♀ 31	(44.3)	♀ 66	(47.5)
Age^b	9.94	(0.46)	9.87	(0.42)	9.90	(0.44)
BMI-SDS^b	1.28	(1.05)	0.86	(1.13)	1.06	(1.11)
IOTF^c,d
Overweight	27	(39.1)	17	(24.3)	44	(31.7)
Obese	6	(8.8)	5	(7.4)	11	(7.9)

IG: intervention group; CG: control group; BMI-SDS: body mass index–standard deviation score; IOTF: International Obesity Task Force; SD: standard deviation.

N = 136.

Values given as N (%) indicate total sample size (percentage).

Values given as M(SD).

Values given as N (%) for both overweight and obese groups of participants. IOTF SD scores cut-offs: overweight = 1; obesity = 2.^d

Obesity Task Force (2012).

Efficacy of a text-messaging programme to promote health behaviours

For each of the outcome behaviours – fruit and vegetable intake, physical activity and screen time – HLM analyses were developed and presented in Table 2. In addition, plots of the estimated growth for the three behaviours in both groups (control and intervention conditions) at pre- (T0) and post-intervention (T1) and follow-up (T1) are presented in Figure 1.

Table 2.

Three-level HLM analysis for the three target behaviours (fruit and vegetable intake, physical activity and screen time) over time (T1, T2 and T3).

Effect	Fruits and vegetables	Physical activity	Screen time
Intercept
Estimate	2.25	1.90	1.53
SE	.18	.15	.18
Significance	<.001	<.001	<.001
Group
Estimate	−0.49	−0.03	0.05
SE	.26	.22	.25
Significance	.083	.141	.853
Time
Estimate	0.08	−0.05	−0.25
SE	.24	.32	.30
Significance	.730	.884	.389
Time-squared
Estimate	−0.08	0.002	0.08
SE	.12	.16	.14
Significance	.469	.992	.564
Group × time
Estimate	0.97	0.82	0.05
SE	.34	.45	.41
Significance	.005*	.068	.897
Group × time-squared
Estimate	−0.33	−0.35	0.30
SE	.16	.22	.20
Significance	.043	.110	.880

HLM: hierarchical linear model; SE: standard error; NS: not significant; CG: control group; IG: intervention group.

Values in ‘Time’ are related to the changes in CG from T1 to T2.

Values in ‘time-squared’ are related to the changes in CG from T2 to T3.

Values in ‘Group’ time’ are related to the changes in IG from T1 to T2.

Values in ‘Group’ time-squared’ are related to the changes in IG from T2 to T3.

p value < .05, n.s. p value ⩾ .05.

Figure 1.

The changes over time in the three target behaviours for both intervention and control groups.

Baseline differences between groups

At baseline (T1), participants reported an average daily consumption of 1.99 portions of fruits and vegetables (SD = 1.04 hours), physical activity of 1.75 hours per day (SD = 1.22 hours) and screen time of 1.53 hours per day (SD = 1.22 hours). There were no significant differences between groups in self-report measures at the baseline, as indicated by the main effect for Group in Table 2 (all p’s ⩾ .083).

Comparison of behavioural changes in intervention group and control group from T1 to T2

Changes in control group

There were no statistically significant changes in terms of fruit and vegetable consumption for the control group (B = .08, standard error (SE) = .24, p = .730) from T1 to T2 (Table 2), as indicated by the main effect for Time. Additionally, no significant changes were observed from T1 to T2 on physical activity (B = −.05, SE = .32, p = .884) and screen time (B = −.25, SE = .30, p = .389) in the control group (see Figure 1).

Differences in changes between intervention and control groups from T1 to T2

The results presented in Table 2 show that the intervention group revealed a significant increase in the intake of fruits and vegetables from T1 to T2 (B = .97, SE = .34, p < .05), compared to the control group as indicated by the Group-by-Time interaction. There were no statistically significant differences between groups from T1 to T2 (see Figure 1) on physical activity or screen time.

Comparison of behavioural changes in both groups from T2 to T3

Change in the control group

For the control group, there were no changes in the trajectory in terms of fruit and vegetables consumption (B = −.08, SE = .12, p = .469), physical activity (B = .002, SE = .16, p = .992) and screen time (B = .08, SE = .14, p = .564) between T2 and T3 (see Figure 1).

Change in the intervention group

From T2 to T3, the intervention group showed a significant reduction in the consumption of fruit and vegetables (B = −.33, SE = .16, p < .05) in comparison to the control group (as indicated by the Group-by-Time-Squared interaction), but still maintaining higher values compared to the control group over time (see Figure 1). Regarding physical activity and screen time, there were no statistically significant differences between groups between T2 and T3 (see Figure 1)

Satisfaction with the monitoring and feedback SMS programme

The majority of the children from the intervention condition, at the end of the monitoring phase, showed they were satisfied with the programme (89.4%) and would recommend it to a friend (78.6%). This group enjoyed the SMS service (72.4%) and thought that the messages were adequate (82.9%). Most participants (87.3%) considered fun the use of the pedometer.

Discussion

This study aimed to test the effectiveness of a SMS-based monitoring and feedback programme to promote health behaviours in children, such as increased fruit and vegetable consumption and physical activity and decreased screen time. The rationale and the design of this study were based on Shapiro et al.’s work (2008). Since text-messaging has become widely used by children, this may be a useful tool for monitoring and improv-ing health behaviours (Shapiro et al., 2008). For example, in Portugal, 82 per cent of the children (9–12 years old) use SMS, and 84.2 per cent own a mobile phone (Cardoso et al., 2007).

Despite the modest outcomes, results revealed that the SMS programme was successful in increasing fruit and vegetable consumption. In particular, the intervention group increased the consumption of fruits and vegetables from T1 to T2, while the control group showed a slight decrease in the intake of this type of healthy food. It seems that the systematic monitoring and the tailored feedback messages could lead to an improvement in this specific behaviour. On the other hand, there were no statistically significant differences between intervention group and control group regarding physical activity and screen time over time. Hence, there is no evidence that the SMS programme was effective in promoting these health behaviours (more activity, less screen time) in the present sample. Only the goal for fruits and vegetables was achieved over time in the intervention group.

However, the baseline data show that this sample revealed high levels of ‘healthy’ behaviour already before the programme, especially in terms of physical activity and screen time. In particular, these children already spent little time in front of a television or computer screen (on average, 1 hour or less per day). Considering that the goal for physical activity was 1 hour per day or 10,000 steps per day, the majority of the participants of this sample were already considered physically active before the intervention. This finding could suggest that the absence of behavioural change over time could be explained by the fact that the participants already achieved the physical activity and screen time goals before the intervention. For this reason, we suggest that the behavioural goals and cut-offs for normative school samples should be adjusted in further studies. For example, setting the goal to 15,000 steps could result in less number of achieved goals in this sample. Hence, children would have received reinforcing messages more frequently and might have tried harder to achieve the goal. Therefore, it seems crucial that the objective for physical activity (10,000 steps per day in this study) should be adjusted in future studies (e.g. 15,000 steps) to investigate the question whether the intervention group would increase its activity levels with reinforcing feedback.

As argued before, monitoring and tailored messages are effective strategies that can have an impact on health behavioural change (Bauer et al., 2010). In addition, text-messaging can be a cost-effective method for increasing the adherence to the intervention and for effecting the behaviour change (Shapiro et al., 2008). A recent study showed, for example, that SMS could be effective tool to deliver maintenance care for overweight and obese children after the intervention (Bauer et al., 2010). In fact, Bauer et al. (2010) concluded that the SMS programme could contribute to the maintenance of the treatment gains: 51.4 per cent of the overweight children maintained or reduced their BMI-SDS during their participation in the SMS programme, introduced after the intervention. However, it is difficult to predict whether or not the behaviour change is maintained long after the use of text-messaging feedback ends. In our study, the group from the intervention condition showed a significant decrease in fruit and vegetable intake from T2 to T3. This decrease occurred during the follow-up period, after the end of the monitoring and feedback. This could mean that, once the intervention is over, the change in the health behaviour might not be sustained. Further studies, to analyse the after-treatment maintenance behaviours, are needed to reinforce the result of this study.

Moreover, considering the baseline data and previous studies (e.g. Bauer et al., 2010; Carter et al., 2012), there is a need to further adapt and apply this type of tools in other population groups. For example, it would be beneficial to test the programme in clinical populations, such as obese children or adults’ samples that want to start a weight loss treatment programme. Thus, the SMS-based programme could be an effective additional tool during the treatment of obese populations (e.g. children). Such a tool could be very helpful for treatment adherence and acceptance and to reduce dropout rates (De Niet et al., 2012).

This study showed that the majority of the participants were satisfied (89.4%) with the programme. As previous research has shown, children prefer technologically tailored and interactive programmes, instead of traditional paper diary or other conservative methods (Shapiro et al., 2008). For this reason, overweight and obese children could benefit from such programmes because, as previous studies showed, self-monitoring can lead to better outcomes (e.g. weight loss) and to greater adherence and acceptance to treatment in health behaviour interventions (Germann et al., 2007; Wei et al., 2011).

Nevertheless, a recent review of mobile text-messaging use in clinical and health behaviour interventions revealed that while the average weekly number of SMS sent was acceptable from weeks 1 to 6, it decreased in the last 2 weeks of the intervention to less than any other day (0.45) (Wei et al., 2011). This result could mean that long periods of self-monitoring and feedback messages could probably lead to disengagement. Possible strategies for keeping children engaged seem necessary, such as sending motivational messages in the last 2 weeks (e.g. ‘Only two weeks to go, you’re almost there!’) and further visits to engage children to complete the programme or reducing the extent of the self-monitoring.

Finally, there were some limitations in this study that should be mentioned. For example, it should be noticed that classrooms were randomized and not the participants individually, which can explain some of the initial differences at baseline levels. In future studies, the participants should be randomized individually, in order to have homogeneous groups at the baseline. Another limitation is related to the self-report instrument used to evaluate the target behaviours in children. This type of measure could raise some issues, especially in the evaluation of physical activity. According to Welk et al. (2000), children are less likely to make accurate self-report assessment, compared to adult populations. On the other hand, self-report instruments are very useful for epidemiological studies or intervention studies where less precision is needed because they require less time and fewer resources when compared with other measures. Nevertheless, in further studies, the use of combined measures (e.g. self-report measures and activity monitors) may be needed to better characterize children’s activity levels (Welk et al., 2000) or other behaviours. Finally, another limitation is the fact that we did not control for the effect of self-monitoring alone, and simply self-monitoring can lead to behaviour change in healthy eating and physical activity (Williams and French, 2011). For this study, only a monitoring plus SMS feedback group was compared to a no monitoring group. Given this limitation, the results of the efficacy of the SMS feedback could not be interpreted separately from the effect of simply self-monitoring. In the future, a randomized controlled efficacy trial could include three groups: SMS monitoring plus feedback, paper–pencil monitoring without feedback and no monitoring.

In conclusion, the present findings suggest that the SMS-based monitoring and feedback systems have potential for promoting health behaviours in children, confirming that text-message can be ‘a potentially powerful tool for behaviour change’ (Cole-Lewis and Kershaw, 2010). Further investigation of such programmes is warranted, especially in a clinical context.

Footnotes

Acknowledgements

We thank Dr Stephanie Bauer and Dr Markus Moessner for their advisory support with the programme, Margarida Silva and Diana Costa for support with the educational sessions, and Lutfi Arikan for the technical assistance. The SMS programme was developed at the Center for Psycho-therapy Research, University Hospital Heidelberg (Germany).

Funding

This research was supported by a Fundação para a Ciência e a Tecnologia/Foundation for Science and Technology, Portugal (FCT/PTDC/PSI-PCL/099981/2008), grant to the last author (P.P.P.M.).

References

Bauer

De Niet

Timman

. (2010) Enhancement of care through self-monitoring and tailored feedback via text messaging and their use in the treatment of childhood overweight. Patient Education and Counseling 79(3): 315–319.

Bauer

Percevic

Okon

. (2003) Use of text messaging in the aftercare of patients with bulimia nervosa. European Eating Disorders Review 11(3): 279–290.

Cardoso

Espanha

Lapa

(2007) E-Generation: Os usos de media pelas crianças e jovens em Portugal [E-Generation: The use of media by children and adolescents in Portugal]. Lisbon: CIES ISCTE.

Carter

Burley

Nykjaer

. (2012) ‘My Meal Mate’: Validation of the diet measures captured on a smartphone application to facilitate weight loss. British Journal of Nutrition 109(3): 539–546.

Cole

(2002) A chart to link child centiles of body mass index, weight and height. European Journal of Clinical Nutrition 56(12): 1194–1199.

Cole-Lewis

Kershaw

(2010) Text messaging as a tool for behavior change in disease prevention and management. Epidemiologic Reviews 32(1): 56–69.

Currie

Gabhainn

Godeau

. (2008) Inequalities in Children’s Health: HBSC: International Report from the 2005/2006 Survey. Geneva: World Health Organization.

De Niet

Timman

Bauer

. (2012) The effect of a short message service maintenance treatment on body mass index and psychological well-being in overweight and obese children: A randomized controlled trial. Pediatric Obesity 7(3): 205–219.

Do Carmo

Dos Santos

Camolas

. (2008) Overweight and obesity in Portugal: National prevalence in 2003–2005. Obesity Reviews: An Official Journal of the International Association for the Study of Obesity 9(1): 11–19.

10.

Germann

Kirschenbaum

Rich

(2007) Child and parental self-monitoring as determinants of success in the treatment of morbid obesity in low-income minority children. Journal of Pediatric Psychology 32(1): 111–121.

11.

Golan

Weizman

(1998) Reliability and validity of the Family Eating and Activity Habits Questionnaire. European Journal of Clinical Nutrition 52(10): 771–777.

12.

International Obesity Task Force (2012) About obesity. Available at: http://www.iaso.org/policy/aboutobesity/ (accessed 19 August 2012).

13.

Lobstein

Baur

Jackson-Leach

(2010) The childhood obesity epidemic. In: Waters

Swinburn

Seidell

. (eds) Preventing Childhood Obesity: Evidence Policy and Practice. Oxford: Wiley-Blackwell, pp. 3–14.

14.

Sardinha

Santos

Vale

. (2010) Prevalence of overweight and obesity among Portuguese youth: A study in a representative sample of 10-18-year-old children and adolescents. International Journal of Pediatric Obesity 6(2-2): e124–e128.

15.

Seidell

(2005) The epidemiology of obesity: A global perspective. In: Crawford

Jeffery

(eds) Obesity Prevention and Public Health. Oxford: Oxford University Press, pp. 3–19.

16.

Shapiro

Bauer

Hamer

. (2008) Use of text messaging for monitoring sugar-sweetened beverages, physical activity and screen time in children: A pilot study. Journal of Nutrition Education and Behavior 40: 385–391.

17.

Tudor-Locke

Hatano

Pangrazi

. (2008) Revisiting ‘how many steps are enough?’ Medicine and Science in Sports and Exercise 40(7 Suppl.): S537–S543.

18.

Tudor-Locke

Pangrazi

Corbin

. (2004) BMI-referenced standards for recommended pedometer-determined steps/day in children. Preventive Medicine 38(6): 857–864.

19.

Uauy

Caleyachetty

Swinburn

(2010) Childhood obesity prevention overview. In: Waters

Swinburn

Seidell

. (eds) Preventing Childhood Obesity: Evidence Policy and Practice. Oxford: Wiley-Blackwell, pp. 22–29.

20.

Vereecken

Maes

(2003) A Belgian study on the reliability and relative validity of the Health Behaviour in School-Aged Children food-frequency questionnaire. Public Health Nutrition 6(6): 581–588.

21.

Wei

Hollin

Kachnowski

(2011) A review of the use of mobile phone text messaging in clinical and healthy behaviour interventions. Journal of Telemedicine and Telecare 17(1): 41–48.

22.

Welk

Corbin

Dale

(2000) Measurement issues in the assessment of physical activity in children. Research Quarterly for Exercise & Sport 71(2 Suppl.): S59–S73.

23.

Williams

French

(2011) What are the most effective intervention techniques for changing physical activity self-efficacy and physical activity behaviour – And are they the same? Health Education Research 26(2): 308–322.

24.

World Health Organization (WHO) (2003) Fruit and Vegetables Promotion Initiative. Geneva: WHO.