Background

Context

Childhood obesity is highly prevalent in the United States (U.S.)¹ and has become a global epidemic.² The recent national survey, the 2007–2008 National Health and Nutrition Examination Survey (NHANES) data showed that 17 percent of U.S. children and adolescents (ages 2–19) years were obese, and over 30 percent were overweight or obese.^3–6 Childhood obesity leads to obesity in adulthood and many other serious health conditions, such as cardiovascular, metabolic, and psychosocial illnesses.²

To assess the effectiveness of existing childhood obesity prevention efforts, the Johns Hopkins University Evidence-based Practice Center completed a systematic review on childhood obesity prevention studies conducted in high-income countries.⁷ This report, funded by the Agency for Healthcare Research and Quality (AHRQ), systematically reviewed seven Key Questions (see Table A).

Table A

Key Questions from “Childhood Obesity Prevention Programs: A Comparative Effectiveness Review and Meta-Analysis”.

The draft Comparative Effectiveness Review evaluated 96 intervention studies reported in 113 articles with the following main conclusions:

• The majority of studies in high income countries are conducted in schools.
• School-based intervention can prevent overweight and obesity, especially those with a home intervention that targets both diet and physical activity.

Though the strength of evidence is moderate to high for school-based interventions, the limited number of studies and insufficient or low strength of evidence to support interventions in other settings made it difficult to conclude that interventions in other settings could effectively prevent childhood obesity. Based on the evidence gaps in these settings, we identified the following as Future Research Needs:

Future research is needed on interventions delivered in settings other than schools or home. Thus, future research is needed for all of the Key Questions except for Key Questions 1 and 2, and especially needed are studies of environmental and policy changes.

While there have been other reviews on the effectiveness of interventions on food and nutrition policies at school on changes in children’s diet and school food environments, there are still gaps in the literature on some aspects, such as the impact of regulations on food availability and its impact on obesity prevention. Only a few studies that we reviewed used social marketing to deliver messages on nutrition, physical activity and health. This approach might be integrated with other intervention components to create an atmosphere favorable to healthy and active lifestyles and related behavioral changes. Additionally, further testing of the value of consumer health informatics products for obesity prevention is needed. In addition, there is a lack of evidence on the impact of regional or national policies on childhood obesity prevention, including agriculture policy and regulations on food retailing and distributions.

Furthermore, further research might be conducted with stratified analyses on subgroups, such as by gender, age, race/ethnicity, or socioeconomic status. This will help us learn how different groups may respond differently to the same intervention, and help tailor future interventions to maximize their benefits.

There were methodological limitations of the reviewed studies which suggest that future research might improve upon the methods. Few of the studies we reviewed reported process evaluation, which would provide useful insight regarding why some studies might detect desirable effect of the intervention. More vigorous analytic approaches are desirable in future studies, to better analyze the repeated measures collected during follow-up, to control for confounders, and to test effect modification.

The studies we reviewed typically had limited followup and we could not know the sustainability of these interventions. Future studies need to design innovative approaches that have a high likelihood of sustainability; for example, studies using a community-based participatory approach. This may be designed to take advantage of other existing public health, government or other organization supported programs or try to gain more support and engagement from related key stakeholders.

The objective of this report is to prioritize the needs for research addressing gaps in the existing literature on the effectiveness of childhood obesity prevention programs by engaging expert stakeholders using a modified Delphi method.

Methods

We identified research gaps from areas of low or insufficient strength in Childhood Obesity Prevention Programs: A Comparative Effectiveness Review and Meta-Analysis.⁷ All Key Questions from that evidence report, except Key Questions 1 and 2, were considered gaps in the literature. We used a modified Delphi process with six expert stakeholders to prioritize individual PICOS elements (populations; interventions, comparisons of interventions; outcomes; settings) to identify Future Research Needs for childhood obesity prevention.

We recruited a variety of stakeholders with potential interest in childhood obesity prevention such as parents, researchers, and representatives from government and public agencies. Stakeholders were recruited via letters, emails, and phone invitations. They were asked to participate in a Delphi process using a Web-based assessment tool. Stakeholders were asked to read the Executive Summary of the 2012 draft evidence report “Childhood Obesity Prevention Programs: A Comparative Effectiveness Review and Meta-Analysis” as a guide to prioritize evidence gaps. Stakeholders were allowed to enter free-text comments during the prioritization process to provide us insight into their thought processes; however, free-text was optional so these comments were not used in prioritization.

Results

During round 1, the majority of the six stakeholders suggested that future studies should enroll children from all age groups (2–18 years), from all income strata or selectively enroll children from low income families. They felt that children of all races or ethnicities will be important to study. Four of six stakeholders regarded a combination of interventions as the priority for future rather than interventions focusing solely on diet or physical activity/sedentary behavior. During round 1, multiple setting or home-based interventions were voted as highest priority by five of six stakeholders.

When the stakeholders were asked in round 2 to select and prioritize combinations consisting of a specified age group, income group, ethnic group, and setting; the stakeholders favored studies of toddlers and young children over adolescents, favored studies of caloric restriction over physical activity interventions, and favored studies conducted at home or in a community setting over studies conducted in schools (were the evidence is sufficient). Given these preferences, we presented the stakeholders with comparative effectiveness questions which they prioritized in round 3. The 12 questions in order of preference are in Table B. The stakeholders prioritized several methodological challenges that we presented to them with the recommendation that these should be addressed in future studies. These included the need for better evaluation of community-based studies, attention to barriers and facilitators of implementation, and improved analytic methods.

Table B

Prioritized research questions.

Each of the study questions described above would best be answered with a randomized control trial (RCT). Depending on the question, recruitment may be from physicians’ offices or public services offices and implementation might be at churches or community centers or may require the involvement of individuals who can make home visits. The interventions that involve the community may benefit from a community-based participatory research framework, and randomization might be at the level of the community. Those interventions targeting Hispanic youngsters will require culturally-appropriate interventions and the availability of Spanish language materials if needed. Scalability will depend on intensity of intervention.

Discussion

Using the 2012 draft evidence report “Childhood Obesity Prevention Programs: A Comparative Effectiveness Review and Meta-Analysis,”⁷ we identified and prioritized Future Research Needs. We identified 12 research questions considered to be of potential health impact by a multidisciplinary group of stakeholders. We expect that this report will help researchers to develop studies evaluating the Key Questions identified, as well as enable funding agencies to dedicate their resources to areas most likely to make a health impact.

Our stakeholders were clear on prioritizing certain groups of the population for further study: the stakeholders favor future research about obesity prevention in toddlers and young children, particularly low income toddlers and children, with attention to Hispanic youngsters. Although the stakeholders acknowledge that additional research is needed for prevention of obesity in adolescents as well, they prioritize research directed at younger children. Similarly, they acknowledge that obesity strikes middle and upper income children and yet recommend that attention be first directed to low income children. Given that the sites of recruitment and the optimal interventions may differ for low income children relative to middle and upper income children, studying this population separately may indeed be necessary.

Our stakeholders greatly favored studying interventions that target caloric restriction over other targeted behaviors. This is not because the evidence already strongly supports or refutes the use of physical activity interventions, but the stakeholders were wary of the efficacy of these interventions particularly for toddlers and young children. They suggested that the settings in which the intervention targeting calorie restriction is delivered is the most pressing question to address: should the intervention be delivered in the home or should it involve a community intervention as well—this community intervention might be as high-level as implementation of legislation regarding food availability (e.g., large soft drinks) or it may be more local such as the implementation of a healthy-eating campaign among local churches. We caution that the interventions must be culturally appropriate for the targeted children, particularly studies that will enroll primarily Hispanic children. The stakeholders do not want to see additional studies conducted in schools at present. They feel that the sufficiency of the evidence makes the other settings greater priorities.

The methodological limitations in the current evidence base should be addressable, and the stakeholders endorsed the methodological challenges that we presented to them. They particularly support the need for improved methods for the evaluation of community based interventions as well as better description (and testing) of barriers and facilitators to implementing proven programs, as well as greater rigor in analyses.

There are some limitations of this project. The large number of evidence gaps made it unfeasible to create and present all research questions from these gaps to our stakeholders, as would be a more standard approach to identifying Future Research Needs. We modified the approach piloted in a prior Future Research Needs report for this purpose.⁸ This method relied heavily on input from the authors of the Comparative Effectiveness Review and the stakeholders, who all have their own priorities and biases that influence their reflections on the Comparative Effectiveness Review process. Additionally, we had hoped to have more stakeholders involved in this process.

There are several strengths to this report. Our research team included several members of the original report’s research team, which provided ready access to their insight on the process of the Comparative Effectiveness Review and challenges experienced by that original team. We also recruited stakeholders to represent a variety of interests. The prevention of childhood obesity is important to not only to clinicians and researchers, but is also of particular interest to parents. We were fortunate to have an engaged parent of an obese child among our stakeholders. We feel that our diverse array of engaged stakeholders helps to ensure that the Key Questions we developed will be of significant public health impact. Finally, we stakeholders were allowed to provide free-text comments in addition to performing rankings. This qualitative component gave insight on thought process behind many of the individual stakeholders’ choice.

Conclusions

Using the 2012 draft evidence report “Childhood Obesity Prevention Programs: A Comparative Effectiveness Review and Meta-Analysis,” we identified and prioritized Future Research Needs. We identified 12 research questions considered to be of potential health impact by a multidisciplinary group of stakeholders. These questions focus on high-priority populations, interventions, comparisons, and settings as identified by our stakeholders. During the systematic review process, we identified methodological issues in the literature; our stakeholders agreed that improving these methods will benefit studies to come. This report may inform and support researchers to develop studies to evaluate the Key Questions identified, as well as enable funding agencies to dedicate their resources to areas most likely to make a health impact.

References

1.

Centers for Disease Control (CDC). Childhood overweight and obesity. Centers for Disease Control (CDC); Updated March 31, 2010. www​.cdc.gov/obesity/childhood/index.html.

2.

Wang Y, Lobstein T. Worldwide trends in childhood overweight and obesity. Int J Pediatr Obes. 2006;1(1):11–25. [PubMed: 17902211]

3.

Ogden CL, Carroll MD, Curtin LR, et al. Prevalence of high body mass index in US children and adolescents, 2007–2008. JAMA. 2010 Jan 20;303(3):242–9. [PubMed: 20071470]

4.

Olson CM, Demment MM, Carling SJ, et al. Associations between mothers’ and their children’s weights at 4 years of age. Child Obes. 2010 Aug 1;6(4):201–7. [PMC free article: PMC3129907] [PubMed: 21743836]

5.

Birch LL, Anzman-Frasca S. Promoting children’s healthy eating in obesogenic environments: Lessons learned from the rat. Physiol Behav. 2011 Sep 26;104(4):641–5. [PMC free article: PMC3139791] [PubMed: 21620880]

6.

Wang Y, Beydoun MA. The obesity epidemic in the United States--gender, age, socioeconomic, racial/ethnic, and geographic characteristics: A systematic review and meta-regression analysis. Epidemiol Rev. 2007;29:6–28. [PubMed: 17510091]

7.

Wang Y, Wu Y, Wilson RF, Showell N, et al. Childhood Obesity Prevention Programs: A Comparative Effectiveness Review and Meta-Analysis. Rockville, MD: Agency for Healthcare Research and Quality; Apr, 2013. (Prepared by the Johns Hopkins University Evidence-based Practice Center under Contract No. 290-2007-10061-I.) AHRQ Publication No. 13-EHC-81-EF. www​.effectivehealthcare​.ahrq.gov/reports/final.cfm. [PubMed: 23865092]

8.

Bennett WL, Nicholson WK, Saldanha IJ, et al. Future Research Needs for the Management of Gestational Diabetes. Future Research Needs Paper No. 7. Rockville, MD: Agency for Healthcare Research and Quality; Nov, 2010. (Prepared by the Johns Hopkins University Evidence-based Practice Center under Contract No. 290-2007-10061-I.) AHRQ Publication No. 11-EHC005-EF. www​.effectivehealthcare​.ahrq.gov/reports/final.cfm. [PubMed: 21433329]