OUP user menu

Changes in BMI and blood pressure after a school based intervention: The CHILDREN study

P. D. Angelopoulos, H. J. Milionis, E. Grammatikaki, G. Moschonis, Y. Manios
DOI: http://dx.doi.org/10.1093/eurpub/ckp004 319-325 First published online: 10 February 2009


Background: Identification of the behavioural, normative and control beliefs influencing children's behaviour is an important prerequisite in designing effective interventions. The current study aims to evaluate the effectiveness of an intervention program, based on the Theory of Planned Behaviour (TPB), on obesity indices and blood pressure (BP) in Ioannina, Greece. Methods: Participants were 646 fifth grade pupils (360 girls and 286 boys). The intervention group (IG) consisted of 321 children in 13 randomly selected schools while the rest (n = 325 children) served as the control group (CG). Based on the outcome of the TPB questionnaire, the intervention focused on overcoming the barriers in accessing physical activity areas, increasing the availability of fruits and vegetables and increasing parental support. General linear mixed model and mediating variable analysis were used to evaluate the differences between the two groups and to test whether changes in certain dietary, physical activity and anthropometrical indices mediated the effect of the intervention on BP.

Results: IG had higher consumption of fruits and lower consumption of fats/oils and sweets/beverages compared with the CG. Intervention's effect on BMI could be explained by the changes in fruit and fats/oils intake whereas the reduction of systolic and diastolic BP could be explained by the reduction of BMI. Conclusions: The findings indicate favourable changes in BP and obesity indices after the implementation of a 1-year school-based intervention program based on the TPB. These results highlight the importance of developing a social and physical environment that promotes balanced eating behaviours and extra-curricular access to physical activity venues.

  • Theory of Planned Behaviour
  • children
  • intervention
  • blood pressure
  • BMI


Many recent reports have indicated that the prevalence of obesity in childhood and adolescence has been increasing worldwide at an alarming rate.1–5 Several prospective studies have indicated that in one-third to one-half of cases, obesity tracks from childhood to adulthood, where it becomes associated with an increased prevalence of cardiovascular disease (CVD).6–8 Next to the tracking phenomenon of obesity, the magnitude of other CVD-related risk factors, such as high blood pressure (BP), has been reported to show the same rising tendency.9–12 In Greece, childhood overweight and obesity is a major problem affecting about 20–30% of children aged 6–17 years old.13,14 However, the prevalence seems to be related to socioeconomic indices thus affecting more the vulnerable populations and in extent poorer regions.15,16 In fact, a closer look in schoolchildren at the age of 11 years old, revealed that overweight and obesity rates at the prefecture of Ioannina—which is one of the poorest in the country—were much higher compared with mean rates in Greece, reaching 29.4 and 11.8% for boys and 39.0 and 7.5% for girls, respectively.12

This epidemic of childhood obesity calls for appropriate measures and drastic actions to be taken early in life whereas early prevention seems as an emerging and effective approach to decrease chronic diseases later in adulthood.17–21 School setting is an ideal place to develop and evaluate innovative health and nutrition education interventions, since children spend much of their time in school. Recent findings from school based intervention programs have delivered some encouraging findings regarding obesity indices indicating the potentiality of such programs.20–25 However, identifying the social and environmental factors that influence children's behaviour and taking them into account when designing the intervention programmes could further improve their effectiveness.26,27

Baring this in mind, a new school based intervention program was designed and implemented. The innovative parts of this new program were that prior to the development and implementation of the programme, a questionnaire based on the Theory of Planned Behaviour (TPB) was structured and implemented in order to identify behavioural, normative and control beliefs, which might restrict or support desired dietary habits and physical activity levels.28

The aim of the current study was to evaluate the effectiveness of this school-based intervention program, which was developed and implemented based on the TPB, on obesity indices and blood pressure in primary school children in the prefecture of Ioannina—a poor area with high obesity rates in Greece.


Sampling and study design

The current study was conducted within Ioannina Metropolitan Area, Greece. There are 51 primary schools (35 urban and 16 rural) consisting of all six different grades (first to sixth) currently operating in the city of Ioannina and in its rural surroundings. A total of 6901 pupils are registered to these primary schools with 1150 of them attending fifth grade. The sampling procedures took into account the distribution of pupils in urban and rural areas as well as the distribution of schools. School was used as the main sampling unit taking into consideration that 73% of pupils in the region attend urban schools and 27% of them attend rural schools.

From the total number of primary schools in the region, a random sample of 26 schools was selected resulting in a total sample of 646 fifth grade pupils (360 girls and 286 boys, participation rate: 96%). Under the following assumptions for a two-sided test: significance level 5%; statistical power 80%, a sample size of 600 individuals would be adequate to detect a BMI difference greater than 1 point or 12% difference in the prevalence of obesity between the two study groups. Three quarters (75%) of these pupils (n = 485) attended urban schools while the rest of them attended rural schools. Before acceptance, children's parents or guardians were fully informed about the objectives and methods of the study and signed a consent form. The study protocol was approved by the Greek Ministry of Education and by the National Educational Institute.

A school-based nutrition and physical activity intervention programme was developed and implemented to a random sample of 13 primary schools (eight urban and five rural) while 13 schools (eight urban and five rural) served as the control group (CG). Schools were randomized to study condition by using a method of random digits. The final population of the intervention group (IG) was 321 children (265 from urban areas and 56 from rural areas), whereas the rest 325 children (242 from urban areas and 83 from rural areas) comprised the CG.

Description of the intervention programme

A 12-month intervention programme (January 2005 to January 2006) was implemented only in IG schools after the approval of the Greek Ministry of Education and was integrated in the existing school curriculum; primarily combined with Physical Education (PE) and Science and Environmental classes providing the less possible disturbance. An intervention material was developed and constituted of a student's workbook and a teacher's manual including on appendix certain activities which could be photocopied, reproduced and used in the class. The themes covered through this manual were self-esteem, body image, nutrition, physical activity, fitness and environmental issues. The material was implemented for 1–2 h per week. Several motivational methods and strategies were used for increasing knowledge (i.e. discussion, active learning, cues), increasing skills and self-efficacy (i.e. modelling, guided practice, enactment), achieving better self-monitoring (i.e. problem solving, goal setting), changing attitudes and beliefs (i.e. self re-evaluation, environmental re-evaluation, arguments, modelling, direct experience) and changing social influence (i.e. modelling, mobilizing social support). The intervention was delivered by the school teachers who were trained by the research team. For the completion of certain activities at home, parental involvement was required to provide the necessary social reinforecement.

The TPB was used to identify the determinants of children's dietary and physical activity behaviours.28 Information regarding the effect of children's behavioural, normative and control beliefs regarding certain health behaviours, namely diet and exercise were obtained from four focus groups during September to October 2004. This information was used for the development of a 92 items questionnaire based on the steps suggested by Azjen and Madden28 and Francis.29 This questionnaire was aiming to assess children's attitudes towards behaviour, subjective norms as well as perceived behavioural control that acted as supporting or restricting factors for their dietary habits or their levels of physical activity. The implementation of the TPB questionnaire revealed that attitudes significantly predicted children's exercise intention, in both boys and girls. From five different attitudes, the strongest predictor of exercise intention for both genders was a child's perception that exercise was fun and exciting. Concerning subjective norms, family support significantly predicted exercise intention in both genders but could not predict exercise behaviour. On the other hand, perceived behavioural controls in the present study were found to be a strong predictor of exercise intentions but also behaviours demonstrating that external and environmental factors can be barriers for low levels of physical activity. In particular, from four different perceived behavioural controls, the stronger negative predictor of exercise intention and behaviour was a child's perception that there was not accessibility to sports facilities and playgrounds.30 Based on these results, special emphasis was placed on increasing children's fun and excitement for exercise. This was achieved by children's participation in two 45-min PE sessions per week (a total of about 60 classes per year) which were delivered in the playground. The sessions were enjoyable, fitness-oriented (rather than motor-oriented) and of moderate intensity. Little attention was placed on competition and winning while verbal rewards were given for all levels of effort and ability.31 In addition, as TPB questionnaire revealed corrective actions should be taken regarding accessibility of playgrounds and parental support. National PE curriculum for Greek schools does not achieve the required levels of motor and cardiovascular health related fitness.32 Therefore, as part of the programme, IG playgrounds and school yards were accessible for children to play after the end of the curricular programme, thus eliminating the factor mainly perceived as restrictive towards engagement in physical activity. In addition, parental support was achieved via meetings at which IG parents were given a file containing the results of their child's medical and nutritional assessment. The parental attendance at these meetings was high with a mean rate of 86%. In addition, parents were advised to support their children in being physically active rather than to encourage sedentary behaviours. Some of the motivational methods used during these meetings were provision of feedback, reinforcement, discussion, persuasive communication, loss/gain frame. CG parents did not attend any sessions and received their child's baseline results and brief comments by mail.

Similarly, TPB questionnaires were applied regarding fruit, vegetables, dairy and sweet consumption. Educational material and motivational intervention implemented promoted the consumption of a balanced diet and thus referred to all these food groups; however for the development of an intervention at the social and physical environment level, it was decided to address only fruit and vegetables. It has been shown that positive messages about what can be eaten (e.g. fruit and vegetables) are more effective than restrictive messages about what not to eat (e.g. fat and sweets).25 Subjective norms, i.e. family and teacher support were shown to positively affect children's intention to consume fruit and vegetable while the strongest negative predictors of fruit and vegetable consumption among the different perceived behavioural controls were limited availability of fruit and vegetables at home and limited access to those at school or neighbourhood setting. Thus the intervention focused on increasing parental involvement and availability of fruits and vegetables at home and school. Parents were encouraged to have more fruit and vegetables at home. Availability at home was assessed periodically as part of activities in the educational material (e.g. ‘Please write down the food items available in your home's fridge and cupboards’ as part of an activity on safe food handling at home). In addition, school canteens were also obliged to have fresh fruit and freshly made juices throughout the whole intervention period. Parents were also involved in the completion of certain educational activities throughout the year whereas fruit or vegetable bazaars were organized, with the participation of the class and the family, to increase familiarity of children with different fruits and vegetables but also to provide the necessary parental or peer support for their consumption. The children attending CG schools could not access the playgrounds or yards of their schools, which is the case for the majority of schools in Greece while they did not receive any support to e.g. overcome the barriers in fruit or vegetable consumption.

From the above-mentioned description of the intervention, it can be elicited that several changes in children's environment (i.e. availability of fruit and vegetable at home, access to playgrounds) were pursued in conjunction with the desired changes in children's behavioural, normative and control beliefs (TPB constructs). This approach reflects the philosophy of the ecological models, which support that the environment is an important factor in shaping various health behaviours and that its effect can be mediated by the TPB variables.33

Evaluation of the intervention

For evaluation purposes of the intervention programme, baseline data collection was carried out from November to December 2004 and follow up data collection from February to March 2006 by a team of trained personnel. The data collected from the children comprised certain physiological and behavioural indices, such as anthropometric, dietary and physical activity indices as mentioned below. The procedures followed are described in more detail elsewhere.12

Anthropometric measurements and blood pressure

Body weight was measured using a digital scale (Seca Personal Floor Scale 861) with an accuracy of 100 g. Subjects were weighed without shoes, in the minimum clothing possible (i.e. underwear). Standing height was also measured without shoes to the nearest 0.1 cm with the use of a portable commercial stadiometer (Leicester Height Measure) with children keeping their shoulders in a relaxed position, their arms hanging freely and their head aligned in Frankfort plane. BMI was calculated by dividing weight (kg) by height squared (m2). The Nutstat module of EpiInfo34 was used to determine the age and sex-specific z-scores for BMI, according to the Centres for Disease Control (CDC) 2000 Growth Charts.35 However BMI z-score was not used in multiple regression analyses since BMI (kg m−2) has been shown to be a more suitable measure of adiposity change in children and adolescents compared with BMI z-score.36,37 BP was measured in the right arm with the subject seated and quiet after a 5-min rest, using a validated automated sphygmomanometer (Omron M5i Blood Pressure Monitor).38

Dietary assessment

The 24-h recall technique was used to collect dietary intake information for two consecutive week days and one weekend day. Study participants were asked to describe the type and amount of food, as well as all beverages consumed during the previous day, provided that it was a usual day according to the participant's perception. To improve the accuracy of food descriptions, standard household measures (cups, tablespoons, etc.) and food models were used to define amounts when appropriate. At the end of each interview, the interviewers, who were rigourously trained to minimize interviewer's effect, reviewed the collected food intake data with the respondent in order to clarify entries, servings and possible forgotten foods. Food intake data were analysed using the Nutritionist V diet analysis software (First Databank, San Bruno, CA, USA), which was extensively amended to include traditional Greek recipes, as described in Food Composition Tables of Greek Cooked Foods and Dishes.39

Physical activity assessment

Physical activity during leisure time was assessed using a standardized questionnaire completed by the children for two consecutive weekdays and one weekend day.40 Reported activities were grouped by a member of the research team into moderate-to-vigorous physical activities (MVPA) (intensity higher than four metabolic equivalents).40

Statistical analysis

Data were reported as Mean (SD) for continuous variables and as mean change [95% Confidence Interval (95% CI)] over baseline. Categorical variables were reported as percentages (%). Differences between proportions were evaluated using chi-square or Fisher's exact test when appropriate. General linear mixed model analysis was used to evaluate the differences between the two groups (i.e. IG vs. CG) with respect to the changes observed over the intervention period. Adjustments were made for gender, school region and BMI wherever appropriate. Inter-school variation was also taken into account by including the random school effect in the models. Finally, a mediating variable analysis was performed in order to test whether changes in certain dietary, physical activity and anthropometrical indices mediated the effect of the intervention on blood pressure.41,42 SPSS 13.0 software (SPSS Inc. Texas, USA) was used to carry out all analyses. All P-values reported were two-tailed. The level of statistical significance was set at P ≤ 0.05.


Table 1 presents certain social and demographical characteristics of the study participants attending the intervention and control schools. According to these data there were no significant differences at baseline between subjects in the intervention and control schools regarding urban–rural split, nationality and parental educational level.

View this table:
Table 1

Social and demographical characteristics of children in the intervention and control schools

Intervention (N = 321)Control (N = 325)P-value
Age (in years)Mean ± SDMean ± SD
10.25 ± 0.4410.29 ± 0.440.335
Gendern (%)n (%)
    Males137 (42.7)149 (45.8)0.428
    Females184 (57.3)176 (54.2)
School region
    Urban265 (82.6)259 (79.7)0.346
    Rural56 (17.4)66 (20.3)
    Greek290 (90.3)286 (88.0)0.347
    Immigrants31 (9.7)39 (12.0)
Paternal educational level
    Primary school23 (7.2)28 (8.6)0.527
    Junior high school81 (25.2)73 (22.4)
    High school130 (40.5)124 (38.2)
    University83 (25.9)92 (28.3)
    Postgraduate training4 (1.2)8 (2.5)
Maternal educational level
    Primary school39 (12.1)36 (11.1)0.399
    Junior high school71 (22.1)76 (23.4)
    High school113 (35.2)130 (40.0)
    University97 (30.2)81 (24.9)
    Postgraduate training1 (0.3)2 (0.6)
  • The significance of the associations between the categorical variables presented in the Table was tested by using chi-square and Fisher's exact test, when appropriate.

Table 2 presents the changes in physical activity and dietary indices in the IG and the CG during the intervention period. Specifically, the increases observed in the IG for time spent to leisure time MVPA were found to differentiate significantly than the decreases observed in the CG. Daily consumption of fruits increased in the IG but decreased in the CG; IG children decreased consumption of total fat/oils and sweets/beverages, while increases were found for CG children. A significant difference was also found in the change observed in the consumption of dairy products between the IG subjects who decreased it and the CG who increased it. No other significant differences were observed.

View this table:
Table 2

Changes in physical activity and dietary indices for the intervention (n = 321) and control group (n = 325)

Mean (SD)Mean (SD)Mean (95% CI)P
Physical Activity indices
    MVPA (min d−1)
        Control47.7 (41.9)31.3 (23.6)−16.4 (−21.1 to −11.7)0.041
        Intervention41.1 (36.6)43.4 (27.2)2.2 (−2.6 to 7.1)
Dietary indices
    Fruit Intake (exchanges d−1)
        Control1.3 (1.5)1.1 (1.6)−0.2 (−0.4 to 0.1)0.044
        Intervention1.1 (1.2)1.5 (1.8)0.4 (0.1 to 0.7)
    Vegetable Intake (exchanges d−1)
        Control1.1 (1.0)1.2 (1.2)0.0 (−0.16 to 0.22)0.680
        Intervention1.2 (1.1)1.0 (1.4)−0.2 (−0.40 to 0.04)
    Dairy Intake (exchanges d−1)
        Control2.8 (1.4)3.0 (1.9)0.2 (0.02 to 0.5)0.008
        Intervention2.7 (1.3)2.5 (1.1)−0.2 (−0.4 to 0.1)
    Fats and Oils Intake (exchanges d−1)
        Control8.7 (5.1)9.4 (5.0)0.7 (0.05 to 1.4)0.028
        Intervention8.0 (4.7)6.4 (3.8)−1.6 (−2.4 to −0.9)
    Meat Intake (exchanges d−1)
        Control5.5 (3.3)4.8 (3.2)−0.7 (−1.2 to −0.2)0.065
        Intervention4.4 (2.5)3.6 (2.9)−0.8 (−1.3 to −0.3)
    Grains Intake (exchanges d−1)
        Control7.4 (3.1)7.7 (3.3)0.3 (−0.2 to 0.7)0.055
        Intervention6.6 (2.7)5.7 (2.8)−0.9 (−1.4 to −0.4)
    Sweets and Beverages Intake (exchanges d−1)
        Control2.6 (2.7)2.8 (3.2)0.2 (−0.2 to 0.6)0.039
        Intervention2.5 (2.2)1.7 (2.2)−0.8 (−1.3 to −0.4)
  • The differences between the two groups were evaluated by using linear mixed models analysis. School was taken as a random effect. Adjustments were made for BMI, gender and school region (i.e. urban/rural).

Table 3 presents the differences between the two study groups in the changes in blood pressure and anthropometric indices induced over the intervention period. Regarding blood pressure, SBP and DBP levels were found to increase in the CG and to decrease in the IG, with the changes differentiating significantly. Regarding BMI the increase observed in the CG was found to differentiate significantly compared to the decrease in the IG.

View this table:
Table 3

Changes in blood pressure and anthropometric indices for the intervention (n = 321) and control group (n = 325).

Mean (SD)MeanMean (95% CI)
Blood pressure indices
    Systolic blood pressure (mmHg)
        Control116.3 (9.2)118.2 (8.3)1.9 (1.0 to 2.9)0.016
        Intervention119.8 (10.3)118.2 (10.1)−1.6 (−2.7 to −0.6)
    Diastolic blood pressure (mmHg)
        Control73.9 (7.9)76.2 (6.9)2.3 (1.6 to 3.1)0.005
        Intervention73.2 (7.9)72.7 (7.8)−0.5 (−1.3 to 0.3)
Anthropometric indices
    Height (cm)
        Control145.2 (8.2)152.9 (8.0)7.6 (7.4 to 7.8)0.696
        Intervention144.1 (9.0)151.8 (8.7)7.8 (7.6 to 8.0)
    Weight (kg)
        Control42.6 (9.2)47.3 (9.3)4.7 (4.5 to 4.9)0.124
        Intervention42.5 (10.2)44.7 (9.2)2.1 (1.9 to 2.4)
    BMI (kg m−2)
        Control20.1 (3.4)20.2 (3.2)0.1 (−0.03 to 0.2)0.047
        Intervention20.3 (3.6)19.2 (2.9)−1.1 (−1.2 to −0.9)
    BMI z-score
        Control0.83 (0.9)0.67 (0.8)−0.16 (−0.19 to −0.12)0.074
        Intervention0.87 (0.9)0.41 (0.9)−0.46 (−0.50 to −0.42)
  • The differences between the two groups were evaluated by using linear mixed models analysis. School was taken as a random effect. All analyses were adjusted for gender and school region (i.e. urban/rural). In the case of blood pressure indices adjustments were also made for BMI.

Table 4 summarizes the results derived from the mediating variable analysis, performed with those variables that were found to differentiate significantly between groups from the mixed model analysis. More specifically, the mediating variable analysis revealed that the effect of the intervention on BMI, SBP and DBP changes was no longer significant (β = –0.08, P = 0.123; β = –0.11, P = 0.065; and β = –0.13, P = 0.053, respectively) after controlling for possible mediators, such as the changes observed in MVPA, fruits’ intake, dairies’ intake, fats’ and oils’ intake, sweets’ and beverages’ intake and BMI. The significant associations between the change observed in BMI and the changes observed in fruit's intake and fats’ and oils’ intake (β = −0.10, P = 0.012 and β = –0.12, P = 0.005, respectively) indicate that the effect of the implemented intervention on BMI as mediated via the changes in fruits’, fat's and oil's intakes. Similarly, the significant associations between the change observed in BMI and the changes in SBP (β = 0.12, P = 0.016) and DBP (β = 0.17, P = 0.001) indicate that the effect of the implemented intervention on SBP and DBP, respectively, was mediated by the change induced in BMI.

View this table:
Table 4

Mediating variable analysis testing for mediators of the relationship between the implemented intervention programme body mass index, systolic and diastolic blood pressure

Model 1Model 2Model 3
Dependent variable: ΔBMI (kg m−2)Dependent variable: ΔSBP (mmHg)Dependent variable: ΔDBP (mmHg)
Independent variablesβP-valueβP-valueβP-value
Treatment arm (0: Control, 1: Intervention)−0.080.123−0.110.065−0.130.053
Δ MVPA (min/day)0.010.853−0.020.7690.0010.987
Δ BMI (kg m−2)0.120.0160.170.001
Δ Dairies’ intake (exchanges d−1)0.050.2450.030.576−0.020.728
Δ Sweets’ and beverages’ intake (exchanges d−1)0.0040.8830.020.730−0.030.603
Δ Fruits’ intake (exchanges d−1)−0.140.005
Δ Fats’ and oils’ intake (exchanges d−1)0.150.002
  • β, Standardized beta coefficient; Δ, denotes the change observed for the specific variable over the intervention period; Adjustments were made for gender and school region (i.e. urban/rural). In Models 2 and 3 Δ fruits’ intake and Δ fats’ and oils’ Intake were not included in the analysis due to their significant association with BMI (Model 1).


In the current study, favourable changes were observed for the consumption of fruits, sweets/beverages and fats/oils but not for vegetables for IG at the end of the intervention period. These findings are in line with those of other school-based programmes aiming to decrease the consumption of sweets and beverages43,44 and fats and oils45 and have resulted in a reduction of overweight and obesity rates. Changes observed in those programmes were achieved with the use of either educational sessions to discourage consumption of the unhealthy food items43,44 or environmental changes (increasing the availability of healthy and decreasing the availability of unhealthy food items at school).45 Baring in mind the results of Slusser et al.46 and Hanson et al.47 who showed that availability at school but also at home are of great importance and need to be addressed when designing an intervention program,46,47 both techniques (educational sessions and environmental changes) were used in the current study to maximize the effect of the intervention. Thus parental facilitation of fruit and vegetables at home was encouraged and the results were mostly favourable for fruits, for which, children's liking is generally higher.48,49 Fruit consumption (including both juice and fruits) rather than vegetable consumption seem to be more likely to change because of differences in availability, convenience or attractiveness.50 However, although mean fruit consumption increased in the IG, a large proportion of the children still consumed less than the recommended servings at the end of the intervention.51

The current study also observed favourable post-intervention physical activity changes for the IG subjects, compared with the CG. The results of the current study can be explained if the low socioeconomic status of the Ioannina prefecture is taken into account. Subjects of low educational level/income or subjects living in neighbourhoods lacking in sports facilities and recreational areas have lower rates of participation in leisure physical activity, and this association seems to exist even in children.52 A possible explanation of the differences observed between the two groups could be the high participation rates in the program but also the diminishing of inequalities in access to leisure time physical activity.53 However, although a post-intervention increase on MVPA was observed for IG children, no association was found between change in time devoted to MVPA and change in BMI in the multiple linear regression model, after controlling for school region and dietary intake. Our findings contradict those of other studies showing that regular physical activity and restriction of sedentarism can help in obesity treatment and reduce BP in children54–56 and those of weight-management trials which consistently report better results when physical activity and/or prevention of sedentary activity are included in the treatment protocol.57 However, other studies have shown that total physical activity may not be influenced by an increase in physical activity after school hours because children compensate for this amount by reducing physical activity in other times of the day.58

Opposing to physical activity, the increase of fruit and decrease of fats and oils consumption was found to be related to the favourable BMI changes observed for the IG children in the multiple regression model. These findings are in accordance with previous studies showing that by decreasing ‘unhealthy’ foods but mostly by increasing ‘healthy’ foods i.e. fruit, significant reductions in BMI can be attained.25,59,60 The regression analysis also showed that the favourable effects observed in the IG pupils for both diastolic and systolic blood pressure could be attributed to the reduction observed in BMI values. Previous studies have also found that fruit consumption and BMI are inversely correlated to blood pressure in both children and adolescents.61–63

In conclusion, the positive findings of the present study are of particular interest considering the scarcity of studies, with respect to the effects of interventions on obesity indices and blood pressure.64–66 The positive changes observed in this study could be mainly attributed to the systematic approach used, as it was based on the results of the TPB questionnaires and could therefore address children's actual needs, attitudes and perceived barriers. The favourable changes in the IG were achieved by the use of measures aiming to reduce the barriers in accessing physical activity areas53 but also by increasing the availability of ‘healthy’ food options. In that context, our results highlight the importance of developing a social and physical environment which encourages and promotes balanced eating behaviours and extra-curricular access to physical activity venues, so that children's BMI and blood pressure indices can be improved.23,67,68 These results can be applicable in other regions in Greece, since the intervention module can be easily incorporated in the existing school curriculum. In addition, given that Ioannina is considered as an economically disadvantaged region of Greece, an increase of the fruit and vegetable availability at home could be easier to be achieved in other more economically robust regions or higher educated populations.69 Taking into consideration the tracking of these indices from childhood to adulthood,70 the inclusion of the above mentioned components in health and nutrition education programmes seems to be an effective tool in tackling chronic disease risk factors early in life.


We thank Prof. Haralampos M. Moutsopoulos, Department of Pathophysiology, School of Medicine, University of Athens, for his guidance and assistance in completing the study. The valuable contribution of Chrisanthy Vlachaki, Zoi Bouloubasi and Marianna Zappi regarding dietary data analysis and Tina Louisa Cook regarding analysis of the TPB questionnaires is also acknowledged.

Conflicts of interest: None declared.

Key points

  • Environmental changes that improved accessibility to playgrounds and schoolyards in the afternoon resulted in an increase of the extracurricular physical activity of children.

  • Increased availability of fruits at home and parental support resulted in positive changes in fruit consumption.

  • An intervention addressing the main behavioural, normative and control beliefs of schoolchildren was found to be successful in lowering their BMI and blood pressure.


View Abstract