From 1980 to 2013, child obesity prevalence increased by 47.1% globally [
Although it has been recognized that schools are ideal settings for obesity prevention initiatives [
The inclusion criteria for systematic reviews such as Cochrane reviews are driven by the participants, interventions, and clinical questions being asked [
A literature search was conducted in PubMed, Health Source, MEDLINE, PsycBOOKS, Psychology and Behavioural Sciences Collection, PsycINFO, SocINDEX, and SPORTDiscus in the years 2007–2016. Various search terms were used including “child overweight,” “child obesity,” “physical activity,” “nutrition,” “health education,” “BMI,” “BMI
Given the widespread use and the impact that BMI may have on government policy [
Intervention duration was classified as short-term (≤6 months), moderate-term (>6 months and ≤12 months), or long-term (>12 months) [
Primary outcomes investigated to determine intervention success included BMI and/or BMI-SDS/
Behavioural moderators included PA, fruit intake, vegetable intake, sedentary time, screen time (including TV viewing time only), and sugar sweetened beverage (SSB) intake. Multiple articles relating to the same study were included if relative outcomes were published separately.
To determine the validity and quality of individual studies, Downs and Black’s validated tool for assessing the methodological quality of randomised and nonrandomised studies of health care interventions was used [
Criteria for assessing study quality and bias are as follows:
Reporting: it included the following points: Is the hypothesis/aim/objective of the study clearly described? Are the main outcomes to be measured clearly described in the introduction or methods section? Are the characteristics of the schools/students included in the study clearly described? Are the interventions of interest clearly described? Are the distributions of principal confounders in each group of subjects to be compared clearly described? Are the main findings of the study clearly described? Does the study provide estimates of the random variability in the data for the main outcomes? Have all important adverse events that may be a consequence of the intervention been reported? Have the characteristics of patients lost to follow-up been described? Have actual probability values been reported for main outcomes except where the probability value is <0.001?
External validity: it included the following points: Were the subjects asked to participate in the study representative of the entire population from which they were recruited? Were those subjects who were prepared to participate representative of the entire population from which they were recruited? Were the staff, places, and facilities, where the patients were treated, representative of the treatment the majority of patients receive?
Internal validity-bias included the following points: Was an attempt made to blind study subjects to the intervention they have received? Was an attempt made to blind those measuring the main outcomes of the intervention? If any of the results of the study were based on “data dredging,” was this made clear? In trials and cohort studies, do the analyses adjust for different lengths of follow-up of patients, or, in case-control studies, is the time period between the intervention and outcome the same for cases and controls? Were the statistical tests used to assess the main outcomes appropriate? Was compliance with the interventions reliable? Were the main outcome measures used accurate (valid and reliable)?
Internal validity-confounding (selection bias) included the following points: Were the patients in different intervention groups (trials and cohort studies) or were the cases and controls (case-control studies) recruited from the same population? Were study subjects in different intervention groups (trials and cohort studies) or were the cases and controls (case-control studies) recruited over the same period of time? Were study subjects randomised to intervention groups? Was the randomised intervention assignment concealed from both patients and health care staff until recruitment was complete and irrevocable? Was there adequate adjustment for confounding in the analyses from which the main findings were drawn? Were losses of patients to follow-up taken into account?
Power: it included the following: Did the study have sufficient power to detect a clinically important effect where the probability value for a difference being due to chance is less than 5%?
A qualitative analysis of the findings was conducted. Similar to Golley and colleagues approach [
See Figure
PRISMA flow diagram of processes for study inclusion.
Table
Summary of study methodology, main findings, and critical appraisal scores.
Location | Study ID (author et al., year, intervention name, and citation number) | Participants ( |
Intervention length (months) | Theory | Primarily delivered by | Study design | Strategies | Intervention characteristics | Critical appraisal score (out of 27) |
---|---|---|---|---|---|---|---|---|---|
Canada, British Columbia | Stock et al., 2007, Healthy Buddies [ |
360 | 10 | N/A | Teacher and older students | NRCT | ED, P | Educational intervention: older students received 45-minute healthy living lesson from the leader weekly; older students paired with younger students and taught the lesson to the younger students in 30-minute sessions, 21 lessons taught, school-wide healthy living theme day. Physical intervention: pairs participated in 30-min aerobic circuits twice a week and were encouraged to engage vigorously | 19 |
|
|||||||||
Chile, |
Kain et al., 2014 [ |
1471 | 12 | N/A | Teachers | RCT | ED, P, and PI | Educational intervention: 8 lessons of HE education for students lasting 90 minutes each. Physical intervention: PE teachers were trained to increased time in PE as well as increased MVPA during PE. Parental involvement: motivational, instructional meetings with parents totaling 45 minutes | 21 |
|
|||||||||
England, Wigan Borough | Fairclough et al., 2013, CHANGE! [ |
318 | 5 | Social Cognitive Theory | Teachers | RCT | ED, PI | Educational intervention: teacher training; weekly lessons lasting 60 minutes of healthy lifestyle curriculum with HE and PA topics. Parental involvement: homework assignments involving parents | 22 |
|
|||||||||
England, northeast | Gorely et al., 2009 GreatFun2Run [ |
589 | 10 | Social Cognitive Theory | Teachers | NRCT | EN, ED, P, and PI | Environmental intervention: local media campaign, running event promotion. Educational intervention: HE and PA components across the curriculum, interactive website. Physical intervention: PE centered around running. Parental involvement: an interactive website was available to parents highlighting the importance of HE and PA, child homework assignments to be completed with parents, information and child PA planner provided | 19 |
|
|||||||||
England, southwest | Kipping et al., 2014, Active for Life Year 5 (AFLY5) [ |
2123 | 6 | Social Cognitive Theory | Teachers | RCT | ED, PI | Educational intervention: 16 lesson plans for PA and HE were delivered over the course of 2-3 school terms. Information on HE and PA was provided to schools intended for newsletters. Parental involvement: interactive homework assignments were given intended to be completed with parents and other family members; information on HE and PA was provided to schools intended for parents | 22 |
|
|||||||||
Greece, Ioannina | Angelopoulos et al., 2009, CHILDREN Study [ |
646 | 12 | Theory of Planned Behaviour | Teachers | RCT | EN, ED, P, and PI | Environmental intervention: access to playgrounds. Fruits and fresh fruit juice were available in the school canteen. |
19 |
|
|||||||||
Netherlands, Rotterdam | Jansen et al., 2011, Lekker fit [ |
2622 | 8 | Theory of Planned Behaviour; Ecological Model of Egger and Swinburn | Teachers/PE teachers | RCT | ED, P, and PI | Educational intervention: health education curriculum including 3 lessons focused on HE, active living, and healthy lifestyle choices. Physical intervention: 3 PE lessons a week taught by a PE teacher; additional activities and sports coordinated with local sports clubs were offered outside school hours on a volunteer basis. Parental involvement: a fitness report card was sent home to parents including child’s weight status, as well as annual health promotion events with local sports clubs | 18 |
|
|||||||||
New Zealand, Otago | Taylor et al., 2007, APPLE [ |
730 | 24 | N/A | Community activity coordinators | NRCT | EN, P | Environmental intervention: installed water coolers; new sport and games equipment being supplied; and discouraged SSBs and encouraged intake of fruits and vegetables. Physical intervention: encouraged PA at lunch, recess, and after school through introduction of new games, sports, and activities. Resources were given to teachers to incorporate activity in class | 19 |
|
|||||||||
New |
Rush et al., 2012, Project Energize [ |
1352 | 23 | N/A | Teachers, supported by exercise and nutrition specialists | RCT | EN, ED, P, and PI | Environmental intervention: specialists promoted active transport, active lunch, and peer leadership of PA outside of school; modifications to the canteen were made to provide healthier snacks; healthy fundraising options were also provided to teachers. Educational intervention: children received classroom lessons on HE during the same 3 weeks their parents attended nutrition sessions. Physical intervention: specialists supported teachers by modelling fundamental movement skills, ball games, fitness activities, and sport games and emphasized keeping all children moving throughout the sessions; teachers were also supported on how to manage children during activity sessions. Parental involvement: 3 information sessions were delivered to parents which included a practical nutrition session | 19 |
|
|||||||||
Norway, southeast | Grydeland et al., 2013, Grydeland et al., 2014, and Bergh et al., 2012, HEIA |
1528 | 20 | Social Ecological Theory | Teachers | RCT | EN, ED, P, and PI | Environmental intervention: active commute to school campaign, new sporting equipment. Educational intervention: classroom lessons about PA and dietary behaviours once a month; classroom posters; and computer programme for 7th graders regarding healthy behaviours. Physical intervention: weekly classroom PA breaks for 10 minutes, training of teachers for PE. Parental involvement: parent information sheet | 23 |
|
|||||||||
Portugal | Rosário et al., 2012 [ |
294 | 6 | Health Promotion Model; Social Cognitive Theory | Teachers | RCT | ED | Educational intervention: nutrition intervention with a PA education component; teachers trained by researchers (36 hours) on intervention delivery and delivery to students (36 hours). 12 lessons included topics such as HE for children, drinking water, fruit and vegetables, foods with low nutritional quality, PA, screen time, and cooking healthy meals | 21 |
|
|||||||||
Spain, Reus, Cambrils, Salou, and Vila-seca | Tarro et al., 2014, Education in Alimentation (EdAl) [ |
1939 | 36 | N/A | Undergraduate medical students | RCT | ED, PI | Educational intervention: 8 healthy lifestyle topics—advancing healthy lifestyles, drinking healthy drinks and avoiding SSBs, improving vegetable and legume intake, eating more fruits and nuts and decreasing high sugar/high fat snacks, promoting healthy habits (PA, home meals), and increasing fruit, dairy, and fish consumption; lessons were delivered in four 1-hour education and activity sessions each year over the course of 3 years for a total of 12 sessions and were not a part of the curriculum; corresponding booklets were used by teachers throughout the year. Parental involvement: similar activities children participated in that were included in the educational booklet children participated in were intended for parents also | 21 |
|
|||||||||
Spain, Granollers | Llargues et al., 2011, Avall [ |
509 | 24 | Investigation, Vision, Action, and Change (IVAC) methodology | Teachers | RCT | EN, ED, and PI | Environmental intervention: sport/games equipment provided to school. Parental involvement: healthy recipes were given to families on a monthly basis as well as literature about local facilities and paths for physical activity | 19 |
|
|||||||||
United States, Louisiana | Williamson et al., 2012, LA Health Project [ |
2060 | 28 | N/A | Teachers + internet programme | RCT | EN, ED | Environmental intervention: modifications were made to cafeteria food and vending machines to increase fruit and vegetables and decrease fat; the PA environmental component included changes to PE curriculum with aims of increasing MVPA to at least 60 minutes/day and decreasing screen time to <2 hours each day. Educational intervention: the environment + education group received an internet based education programme and classroom instruction during class time | 21 |
|
|||||||||
United States, South Dakota | Story et al., 2012, Bright Start [ |
454 | 24 | N/A | Only environmental | RCT | EN, P, and PI | Environmental intervention: modifications were made to cafeteria offerings of low-fat foods, low-fat nonflavoured milk, serve recommended portion sizes, increase fruit and vegetable availability and only gave 2nd helpings of fruit and vegetables; teachers limited snacks in the classroom to low-fat and low-sugar foods, and children were encouraged to drink water instead of SSBs. Physical intervention: each child received 60 minutes of PA each day during school hours (walks outside, modifications to PE, active classroom breaks, and active recess). Parental involvement: home environment, family nights, newsletters, and motivational telephone calls | 22 |
RCT = randomised controlled trial.
NRCT = nonrandomised controlled trial.
EN = environmental.
ED = educational.
P = physical.
PA = physical activity.
HE = healthy eating.
PI = parental involvement.
Table
Summary of primary outcomes and moderator variables.
Location | Study ID (author et al., year, and intervention name) | Primary measures (growth reference) | Moderator variables | Primary outcomes | Moderator outcomes |
---|---|---|---|---|---|
Canada, British Columbia | Stock et al., 2007, Healthy Buddies [ |
BMI | X | Younger group: no change; older group: positive | X |
|
|||||
Chile, Ñuñoa | Kain et al., 2014 [ |
BMI |
MVPA (p) | positive | Positive |
|
|||||
England, Wigan Borough | Fairclough et al., 2013, CHANGE! [ |
BMI, BMI |
PA and sedentary time (a), dietary intake (24-hour recall food intake questionnaire) | BMI, no change, BMI |
Sedentary time, no change, light PA, positive, moderate PA, no change, vigorous PA, no change, and fruit and vegetable intake, no change |
|
|||||
England, northeast | Gorely et al., 2009 GreatFun2Run [ |
BMI, BMI-SDS (UK 1990) | PA (a, p), fruit and vegetable intake (24-hour recall with interview) | BMI, positive, BMI-SDS, positive | MVPA, positive, fruit and vegetable intake, no change |
|
|||||
England, southwest | Kipping et al., 2014, Active for Life Year 5 (AFLY5) [ |
BMI | MVPA and sedentary time (a), screen time (q), fruit and vegetable consumption (q), and high energy drink intake (q) | No change | Weekend screen time, positive, high energy drinks, positive, weekday screen time, no change, fruit and vegetable consumption, no change, MVPA, no change, and sedentary time, no change |
|
|||||
Greece, Ioannina | Angelopoulos et al., 2009, CHILDREN Study [ |
BMI, BMI |
dietary intake (24-hr recall with interview), PA (q) | BMI, positive, BMI |
MVPA, positive, fruit intake, positive, SSBs, positive, and vegetable intake, no change |
|
|||||
Netherlands, Rotterdam | Jansen et al., 2011, Lekker fit [ |
BMI, BMI-SDS (IOTF) | X | BMI, no change, BMI-SDS, no change | X |
|
|||||
New Zealand, Otago | Taylor et al., 2007, APPLE [ |
BMI |
Dietary intake (Short Food Questionnaire), PA (a), and PA and television viewing time (Physical Activity Questionnaire for Older Children) | Positive | Carbonated beverages, positive, fruit intake, positive, vegetable intake, no change, higher accelerometer counts at year 1, positive, accelerometer counts, no change, PA (q), negative, and TV viewing time, no change |
|
|||||
New Zealand, Waikato | Rush et al., 2012, Project Energize [ |
BMI-SDS (UK 1990) | X | No change | X |
|
|||||
Norway, southeast | Grydeland et al., 2013, Grydeland et al., 2014, and Bergh et al., 2012, HEIA [ |
BMI, BMI |
PA (a) | Total group: BMI, no change, BMI |
PA, positive |
|
|||||
Portugal | Rosário et al., 2012 [ |
BMI | Dietary intake (24-hr recall with interview), PA (q) | No change | Vegetable intake, positive, fruit intake, positive, and PA, no change |
|
|||||
Spain, Reus, Cambrils, Salou, and Vila-seca | Tarro et al., 2014, Education in Alimentation (EdAl) [ |
BMI, BMI |
Eating habits (self-report), after-school PA (q) | BMI, no change, BMI |
Fruit and vegetable intake, no change, after-school PA in participants who engaged in >5 hours per week at baseline, positive |
|
|||||
Spain, Granollers | Llargues et al., 2011, Avall [ |
BMI | Eating habits (FFQ and Krece Plus test), PA (q) | Positive | fruit intake, positive, vegetable intake, no change, SSBs, no change, and PA, positive |
|
|||||
United States, Louisiana | Williamson et al., 2012, LA Health Project [ |
BMI |
School food selection and intake (digital photography), PA, and sedentary time (Self-Administered PA Checklist) | PP total: no change; PP overweight: no change; PP + SP total: no change; PP + SP overweight: no change; EM total: no change; EM overweight: no change; and EM white girls: positive | PP: PA, no change, sedentary time, no change; PP overweight: PA, negative; PP + SP: PA, no change, sedentary time, no change; PP + SP overweight: PA, no change; and EM: PA, no change, sedentary time, no change |
|
|||||
United States, South Dakota | Story et al., 2012, Bright Start [ |
BMI, BMI |
X | BMI, no change, BMI |
X |
HB = health behaviour.
HK = health knowledge.
HA = health attitudes.
a = accelerometer.
p = pedometer.
q = questionnaire.
PA = physical activity.
MVPA = moderate vigorous physical activity.
HE = healthy eating.
SSB = sugar sweetened beverages.
WHO = World Health Organisation.
IOTF = international obesity task force.
UK 1990 = United Kingdom 1990.
CDC = center for disease control.
PP = primary prevention.
PP + SP = primary prevention + secondary prevention.
EM = environmental modification.
Ten studies measured total PA or moderate and vigorous PA (MVPA), and 4 of these studies resulted in improvements in these variables alongside improvement in BMI in the whole intervention group [
One study reported follow-up measures once the intervention ceased [
The findings of the current systematised review suggest that school-based interventions that include HE and PA components are moderately effective methods for improving BMI in elementary school children which is consistent with the findings of others [
Only one study reported age differences [
Moderators for BMI improvement included increased PA, lowered SSB intake, and increased fruit intake. The studies in this review that measured sedentary behaviour and screen time did not result in reductions in these behaviours or improvements in BMI. This is in contrast to DeMattia and colleagues’ review that found that two of the three included elementary school studies were effective in reducing sedentary behaviours with one noting improvements in BMI [
PA and/or MVPA was the most reported moderator with six studies using objective measures (accelerometer or pedometer) and five studies using questionnaires. In those studies that captured objectively measured PA, three studies demonstrated improvements in PA alongside improvements in BMI. Reduction of SSB intake has been reported by parents and children to be one of the easiest health behaviours to modify [
None of the studies measuring vegetable intake increased this variable alongside BMI improvement. Increasing vegetable consumption appeared more difficult than increasing fruit intake which may be attributed to the child’s perception of fruit being more palatable than vegetables [
Teachers play a strong role in a child’s social environment and have the potential to positively influence behaviours through environmental and social interactions [
Multiple reviews have stressed the importance of basing child obesity interventions on behaviour change theories [
It was unclear how parental involvement influenced intervention effectiveness given the disparity between levels of parental involvement across studies. Six of the eleven studies that included a parental involvement component resulted in BMI improvement; however, three of the four studies that did not include a parental involvement component within their study design also reported improvements in BMI. These findings are in line with Cook-Cottone and colleagues review which found that interventions with minimal or moderate degrees of parental involvement achieved similar BMI results to those without a parental involvement component [
While multiple combinations of environmental, educational, and physical strategies demonstrated the capacity to improve child BMI, education-only interventions may not be sufficient to induce behaviour change. In line with SCT, our findings suggest that if a child’s environment does not support and reinforce new knowledge and attitudes from education and/or the child does not practice the new PA knowledge through performing PA in a supportive environment, the likelihood of inducing behaviour change may be low [
A number of limitations must be considered. Methodological limitations included the absence of quantitative assessment, use of one reviewer, and the use of BMI as an obesity marker. Although a qualitative review by one reviewer limited the type of conclusions that could be drawn, this review’s focused approach allowed for a detailed synthesis of the most widely used obesity indicator in school-based interventions. Additionally, the use of other obesity measures such as waist circumference, body fat %, or waist-to-height ratio may give a better representation of child disease risk [
Methodological limitations of studies included a lack of reporting of adverse events, reporting characteristics of participants lost to follow-up, blinding subjects and assessors to conditions, measuring intervention fidelity, concealing intervention assignment from schools, reporting if participants who agreed to take part were representative of the population, and taking participants lost to follow-up into account. Thorough reporting procedures and the control for biases that threaten internal validity will allow the reader to make a fair judgment of study findings. Although great effort is required to carry out high-quality studies in school-based interventions, it is possible that publication bias in terms of researchers not reporting negative findings in studies may have influenced the results of this review.
Findings from this systematised review suggest that long-term initiatives that include a parental component and involve multiple environmental, educational, and physical strategies may be the most promising for improving indices of adiposity in elementary school aged children. Future school-based interventions designed to improve children’s weight status should focus efforts to increase PA, decrease sedentary behaviours, lower SSB intake, and increase fruit intake, as well as BMI improvement. Targeted moderators could include increasing PA, lowering SSB intake, and increasing fruit intake. Although it is unlikely that one specific school-based intervention can be effective across different cultures, the identification of these moderators that have demonstrated promise should be incorporated into future efforts in combating the perpetuation of child obesity.
The authors declare that they have no competing interests.