Table 2.
Study Characteristics of 13 Studies Included in the Systematic Review, Organized by Primary Outcome, Results, and Alphabetized by Study Name; Data from Published Peer-Reviewed Literature and Reports
| Citations | Location | Study name (years) | Research/evaluation design | Sample size (intervention, control) | Target age (baseline mean ± SD) | Intervention length | Intervention description | Study phases with participatory community engagement26 | Coalition description | Results |
|---|---|---|---|---|---|---|---|---|---|---|
| Primary outcome: anthropometrics | ||||||||||
| Positive results | ||||||||||
| Bell et al.51 and Sanigorski et al.38 | Barwon South West region, Victoria, Australia | Be Active Eat Well (2002–2006) | Nonrandomized controlled design | 1001, 1183 (consented to participate) | 4–12 years Intervention: 8.2 ± 2.3 years Control: 8.3 ± 2.2 years |
3 years | Capacity building intervention to promote healthy eating, physical activity, and decreased screen time in schools (e.g., nutrition policies, walking school buses) and in the community (e.g., social marketing, community events, restaurants) | □ Problem identification ☑ Design and planning ☑ Implementation ☑ Evaluation □ Dissemination ☑ Sustainability |
Stakeholders supported capacity building process and served on various groups (Steering, Local Implementation, Management, and Reference Committees).51 Representation included healthcare, local government, and a local neighborhood renewal organization38 | Relative to controls, BMI z-score significantly decreased by 0.11 units in the intervention group from baseline to follow-up38 |
| Chomitz et al.28 | Cambridge, Massachusetts, United States | Healthy Living Cambridge Kids (2003–2007) | Pre-post design with longitudinal cohort |
Eligible: 3561, N/A (school data, no consent) Analytic sample: 1858, N/A |
K-5th grade (∼5–11 years) 7.7 ± 1.8 years |
3 years | CBPR “grassroots” environmental and policy intervention to promote child healthy weight and fitness in schools (e.g., expanded physical education, gardens), at home (e.g., annual BMI and fitness reports), and in the community (e.g., city-wide policies, family nights) | ☑ Problem identification ☑ Design and planning ☑ Implementation ☑ Evaluation ☑ Dissemination ☑ Sustainability |
Healthy Children Task Force representation included researchers, schools, healthcare, public health, local government, community organizations, and parents. The task force convened in 1990 to promote children's health and began prioritizing healthy eating, healthy weights, physical activity, and fitness in 2000. Members were involved in formative work (e.g., fundraising, piloting interventions) and stayed engaged in all other phases of the research. | BMI z-score significantly decreased by 0.04 units from baseline to follow-up |
| Martinie et al.59 and Benjamin Neelon et al.27 | Mebane, North Carolina, United States | Mebane on the Move (2011–2012) | Pre-post natural experiment with comparison community | 64, 40 (consented to participate) | 5–11 years Intervention: 7.8 ± 1.8 years Control: 8.3 ± 1.9 years |
6 months | Environmental change intervention to promote physical activity before and after school (e.g., walking and running clubs), at home (e.g., portable play equipment), and in the community (e.g., installation of sidewalks, crosswalks, and walking trails) | ☑ Problem identification ☑ Design and planning ☑ Implementation □ Evaluation □ Dissemination ☑ Sustainability |
The community-driven intervention was funded, planned, and implemented by a group of community members with representation from local business, schools, government, and health sectors. The group approached university researchers to be an evaluation partner27,59 | Relative to controls, BMI z-score significantly decreased by 0.5 units in the intervention group from baseline to follow-up.27 |
| Economos et al.29,52 | Greater Boston area, Massachusetts, United States | Shape Up Somerville: Eat Smart, Play Hard™ (2002–2005) | Nonrandomized controlled design | 631, 1065 (consented to participate) | 1st–3rd grade (∼6–9 years) Total sample: 7.6 ± 1.0 years |
2 school years | CBPR environmental change intervention to promote healthy eating and physical activity access and availability before, during, and after school, at home, and in the community (e.g., restaurants, media) | ☑ Problem identification ☑ Design and planning ☑ Implementation ☑ Evaluation ☑ Dissemination ☑ Sustainability |
Community Advisory Council membership was informed by existing relationships and by engaging community members in formative work (focus groups and key informant interviews). Multiple sectors were involved in intervention implementation (e.g., parents, teachers, foodservice, local government, healthcare)29 | Relative to controls, BMI z-score significantly decreased by 0.10 units in the intervention community from baseline to year 1 follow-up29 and by 0.06 units from baseline to year 2 follow-up52 |
| Mixed results | ||||||||||
| de Groot et al.53 and de Silva-Sanigorski et al.36 | Geelong, Victoria, Australia | Romp & Chomp (2004–2008) | Nonrandomized controlled design with pre-post cross-sectional evaluation |
2 year-old-sample: Pre: 1587, 17,732 Post: 1611, 21,911 3.5 year-old-sample: Pre: 1191, 14,647 Post: 1239, 19,050 (data from state health service) |
0–5 years Evaluation: 2 and 3.5 years |
4 years | Capacity building53 and environmental (policy, sociocultural, physical) change intervention to promote healthy eating and active play in early education and care settings | ☑ Problem identification ☑ Design and planning ☑ Implementation □ Evaluation □ Dissemination ☑ Sustainability |
Community stakeholders were consulted on intervention development and a management committee supervised implementation. The committee represented academia, healthcare, local government, early education, recreation, local health department, and oral health.36 |
2 year-old-sample: Relative to controls, no significant intervention effect on BMI z-score. 3.5 year-old-sample: Relative to controls, BMI z-score significantly decreased by 0.04 units in the intervention group from baseline to follow-up.36 |
| Null results | ||||||||||
| Pettman et al.37,54 | Greater Adelaide area, South Australia, Australia | Eat Well Be Active Community Programs (2006–2009) | Nonrandomized controlled design with pre-post cross-sectional evaluation |
4–5 year-old-sample: Pre: 464, 541 Post: 455, 789 (data from state health service) 10–12 year-old-sample: Pre: 836, 790 Post: 590, 608 (consented, with measures) |
0–18 years Evaluation: 4–5 and 10–12 years |
3 years | Community development and capacity building intervention to promote healthy eating and physical activity via 6 strategies (professional development, policy, infrastructure, programs and resources, promotion, and community development) across multiple settings (e.g., child care, health, recreation) | ☑ Problem identification ☑ Design and planning ☑ Implementation ☑ Evaluation □ Dissemination ☑ Sustainability |
3 age-specific Local Action Groups in each intervention community (under 5 s, primary school, youth) gave input on local contexts and ideas for intervention components54 | Relative to controls, no significant intervention effect on BMI z-score in either age group37 |
| Gantner and Olson55 and Olson et al.32 | Upstate New York State, United States | Healthy Start Partnership (2005–2009) | Nonrandomized controlled design with two nonconcurrent cohorts (controls recruited first and not exposed to intervention) |
Recruited during pregnancy: 276, 219 Infant analytic sample: 257, 207 |
0–7 months (mean not reported) | 4 years | Community-based environmental change interventions to promote healthy pregnancy weight gain, postpartum weight loss, and infant growth. Interventions were delivered within and across counties and largely focused on the social environment (e.g., social marketing campaign). Targeted behaviors included healthy eating, breastfeeding, and physical activity | □ Problem identification ☑ Design and planning ☑ Implementation □ Evaluation □ Dissemination ☑ Sustainability |
The Healthy Start Partnership was convened by university researchers in 2005. Representation included nutrition, public health, maternal and child health, hospitals, and community-based organizations with 58 total partners (majority mid-level professionals). Partners met by county and regionally to exchange information and build capacity to develop and implement obesity prevention interventions. | Relative to controls, no significant intervention effect on infant rapid weight gain (defined by weight-for-length z-score trajectories)32 |
| de la Torre et al.56 and Sadeghi et al.33 | Central Valley, California, United States | Niños Sanos, Familia Sana (2011–2016) | Quasi-experimental controlled design (communities randomly selected after key stakeholders agreed to participate) | 378, 217 (consented, with baseline measures) | 3–8 years Intervention: 6.1 ± 1.3 years Control: 6.1 ± 1.2 years |
3 years | CBPR behavioral intervention among Mexican-origin children and their families to promote healthy eating and physical activity in schools and in the community (e.g., Family Nights, community events). Families received monthly vouchers to purchase fruits and vegetables | □ Problem identification ☑ Design and planning ☑ Implementation □ Evaluation □ Dissemination □ Sustainability |
University researchers convened Community Advisory Councils in both intervention and control communities to monitor progress and give input on implementation factors.33 The councils met quarterly and included representation from local government, schools, faith-based organizations, and healthcare (including community health workers, or promotores, to aid with recruitment and implementation)56 | Year 1 results: Relative to controls, no significant intervention effect on BMI z-score33 (results from years 2–3 were not published at the time of this review) |
| Gomez et al.57,a and Gomez Santos et al.35 | 10 municipalities, Spain | Programa Thao-Salud Infantil (2007–2011) | Pre-post design with longitudinal cohort | Analytic sample: 6697, N/A (eligible sample size not reported) | 3–12 years (mean not reported) | 4 years | EPODE49 community-based intervention to promote healthy lifestyles among children and families (i.e., healthy eating, physical activity, adequate sleep, and quality of life). The intervention includes multiple health promotion activities in multiple municipal settings (e.g., schools, food markets, healthcare) | □ Problem identification □ Design and planning ☑ Implementation □ Evaluation □ Dissemination ☑ Sustainability |
In each municipality, program implementation was led by a multidisciplinary team (Thao local team) with stakeholder representation from local government, healthcare, schools, sports, libraries, social services, media, and food markets. Local project managers were selected by local government leaders57 | From baseline to follow-up, prevalence of overweight and obesity increased from 27.3% to 28.3%.35 (statistical significance not reported) |
| Eisenmann et al.60 and Gentile et al.30 | Lakeville, Minnesota and Cedar Rapids, Iowa, United States | Switch® what you Do, View, and Chew (2005–2006) | Randomized controlled design with pre, post, and 6-month follow-up | 670, 653 (consented to participate) | 3rd–5th grade (∼8–11 years) Total sample: 9.6 ± 0.9 years |
8 months | Behavioral and environmental change intervention to decrease screen time (primary behavioral outcome) and promote physical activity and fruit and vegetable intakes in schools, among families, and in the community (e.g., messaging campaign to build public awareness) | □ Problem identification ☑ Design and planning ☑ Implementation □ Evaluation □ Dissemination ☑ Sustainability |
In 2004, a coalition was convened “to give the project high visibility, and to advocate for and sustain the project (page 6).” Representation included community leaders from school, healthcare, government, business, and religious sectors60 | Relative to controls, no significant intervention effect on BMI z-score30 |
| Negative results | ||||||||||
| De Henauw et al.39 and Pigeot et al.58 | Belgium, Cyprus, Estonia, Germany, Hungary, Italy, Spain, and Sweden | IDEFICS (2007–2010) | Nonrandomized, controlled design in each country | 8482, 7746 (consented, with no missing data) | 2–9.9 years Intervention: 6.1 ± 1.8 years Control: 6.0 ± 1.8 years |
2 years | Promoted healthy diet, reduced screen time, increased physical activity, more family time, and sleep duration in schools (e.g., installation of water fountains), at home (e.g., messaging and education), and in the community (e.g., infrastructure changes to promote safe outdoor play). Interventions developed centrally and then adapted and translated for each community58 | □ Problem identification □ Design and planning ☑ Implementation □ Evaluation □ Dissemination □ Sustainability |
In each community, local program managers worked with local intervention program committees to implement the intervention. Implementation was supported by community platforms (representation from local government, public health, and other health stakeholders) and school platforms (representation from teachers and parents).58 | Prevalence of overweight and obesity increased in intervention (19.0–23.6%) and control (18.0–22.9%) groups from baseline to follow-up. No significant difference between groups39 |
| Gombosi et al.31 | North-central Pennsylvania, United States | Tioga County Fit for Life Project (1999–2004) | Nonrandomized controlled design | Eligible: 4804, not reported (school data, no consent) | K-8th grade (5–14 years) (mean not reported) | 5 years | Multicomponent intervention to promote nutrition and physical activity in schools, among families (e.g., occupational health fairs), in the broader community (e.g., Family Fun Days), and healthy eating in restaurants | ☑ Problem identification ☑ Design and planning □ Implementation □ Evaluation □ Dissemination □ Sustainability |
The Fit for Life task force formed in 1997 and served as an advisory board for the intervention. Representation included healthcare, public schools, and private schools. The task force developed goals and helped plan the school-based intervention curriculum | Prevalence of overweight and obesity increased across all grades (intervention effect relative to control group not reported). |
| Primary outcome: behavior | ||||||||||
| Positive results | ||||||||||
| Martinie et al.59 and Benjamin Neelon et al.27 | Mebane, North Carolina, United States | Mebane on the Move (2011–2012) | Pre-post natural experiment with comparison community | 64, 40 (consented to participate) | 5–11 years Intervention: 7.8 ± 1.8 years Control: 8.3 ± 1.9 years |
6 months | Environmental change intervention to promote physical activity before and after school (e.g., walking and running clubs), at home (e.g., portable play equipment), and in the community (e.g., installation of sidewalks, crosswalks, and walking trails) | ☑ Problem identification ☑ Design and planning ☑ Implementation □ Evaluation □ Dissemination ☑ Sustainability |
The community-driven intervention was funded, planned, and implemented by a group of community members with representation from local business, schools, government, and health sectors. The group approached university researchers to be an evaluation partner27,59 | Relative to controls, intervention children significantly increased moderate-to-vigorous and vigorous physical activity by 1.3 and 0.8 minutes/hour, respectively, from baseline to follow-up (assessed by accelerometry)27 |
| Mixed results | ||||||||||
| Eisenmann et al.60 and Gentile et al.30 | Lakeville, Minnesota and Cedar Rapids, Iowa, United States | Switch what you Do, View, and Chew (2005–2006) | Randomized controlled design with pre, post, and 6-month follow-up | 670, 653 (consented to participate) | 3rd–5th grade (∼8–11 years) Total sample: 9.6 ± 0.9 years |
8 months | Behavioral and environmental change intervention to decrease screen time (primary behavioral outcome) and promote physical activity and fruit and vegetable intakes in schools, among families, and in the community (e.g., messaging campaign to build public awareness) | □ Problem identification ☑ Design and planning ☑ Implementation □ Evaluation □ Dissemination ☑ Sustainability |
In 2004, a coalition was convened “to give the project high visibility, and to advocate for and sustain the project (page 6).” Representation included community leaders from school, healthcare, government, business, and religious sectors60 | Relative to controls, intervention children had significantly lower parent-reported screen time postintervention (Cohen's d = 1.26) and at 6-months follow-up (d = 1.38) (effect size ∼2 hours/week). Child-reported screen time findings were not statistically significant30 |
| Primary outcome: environment and policy | ||||||||||
| Positive results | ||||||||||
| Subica et al.34,61 | Communities in 16 states, United States | Communities Creating Healthy Environments Initiative (2009–2014) | Postintervention evaluation | 21 organizations and tribal nations representing communities of color (grantees), N/A | All children (age not specified) | 3 years | Community organizing health promotion interventions to advocate for environment and policy change related to children's healthy food and safe recreation access among African American, Hispanic/Latino, Asian American, and American Indian/Alaskan Native communities | ☑ Problem identification ☑ Design and planning ☑ Implementation □ Evaluation □ Dissemination ☑ Sustainability |
Grantees raised issue awareness and advocated for change by mobilizing coalitions, creating new stakeholder alliances and partnerships, and building local capacity and leadership through youth and resident engagement | 72 policy winsb (per grantee, mean 3.43 ± 1.78; range 1–8) related to healthy food access, recreation, healthcare, clean environments, affordable housing, and safe neighborhoods34 |
The article narrative includes citations for studies' primary outcomes paper, whereas this table includes additional citations if applicable (e.g., process evaluation article that includes information about the coalition's involvement). Mebane on the Move and Switch what you Do, View, and Chew had primary anthropometric and behavioral outcomes and therefore these studies are presented twice in the table; study characteristics (e.g., location, intervention description, and coalition description) are included in both anthropometric and behavior sections for ease of interpretation.
a
Protocol for a separate 2-year school-based evaluation of Programa Thao-Salud Infantil that includes information about the coalition.
b
Policy wins definition: “concrete, quantifiable movements toward improving local environments and policies to support children's healthy eating and active living that were directly traceable to grantees' CCHE interventions and obtained during or within 6 months postintervention.” (page 918).34
EPODE, Ensémble Prevenons l'Obesité des Enfants (Together Let's Prevent Childhood Obesity); IDEFICS, Identification and prevention of Dietary- and lifestyle-induced health EFects In Children and infantS; Niños Sanos, Familia Sana: Healthy Children, Healthy Family; Programa Thao-Salud Infantil: Thao-Child Health Programme; N/A, not applicable.