Background
In response to the growing prevalence of childhood obesity, the 2010 Patient Protection and Affordable Care Act promotes the utilization of preventive care, mandating that insurers cover obesity screening and counseling services as part of insurance policies [
1]. In 2007, a U.S. expert committee recommended that obesity counseling be conducted in the primary care setting, specifically suggesting that pediatricians use patient-centered techniques such as motivational interviewing to counsel patients about improving health behaviors [
2]. The use of motivational interviewing techniques in clinical settings has been reported to be programmatically feasible and effective in improving health behaviors [
3,
4], but it is also important to assess the time and financial resources needed to implement such programs. Decision makers such as health providers, health care payers, and policymakers can use analyses of the costs of health interventions to assess the affordability of interventions and to prioritize resources among competing programs.
Only a dozen analyses in the peer-reviewed literature detail the costs of delivering childhood obesity interventions [
5‐
16]. Just one study, the Australian Live, Eat, and Play (LEAP) intervention, has supported evaluations of the cost of a primary care-based obesity intervention trial [
9‐
11]. The cost of LEAP varies between studies, so we cannot draw definitive conclusions about the affordability of a primary care-based obesity prevention program from those analyses. No analyses have evaluated the cost of primary care-based obesity counseling in the context of the U.S. health care system.
High Five for Kids was an obesity prevention program developed for 2-6.9 year old children who were obese, or overweight with at least one overweight parent [
4]. The intervention has been found to be effective in decreasing television viewing and decreasing unhealthy eating practices among young children [
4]. The aim of this study was to assess the costs associated with implementing the High Five for Kids intervention measured against usual care practices.
Discussion
Intervention costs were higher than usual care costs in our base case and sensitivity analyses. From a societal perspective, the cost of High Five for Kids was $259 per child in the base case analysis compared to $63 per child in usual care, for a net cost of $196 per child (95% CI [$191, $202]). Provider-incurred costs were the primary driver of intervention costs. The cost of the intervention was higher in sensitivity analyses that considered the inclusion of training costs and costs calculated from the perspective of the health plan.
This analysis of the cost of High Five for Kids quantifies the economic burden of one primary care-based motivational interviewing intervention. Data on training costs and direct medical costs may be of particular interest to providers seeking to implement such an intervention, and information on costs from the perspective of the health plan may be of particular interest to payers. In order to get a better sense of the value of the intervention—the gains in health relative to the dollars spent on care—a cost-effectiveness analysis should be conducted. High Five for Kids measured program effectiveness in terms of changes in body mass index (BMI, kg/m
2) and BMI z-score [
4], but it is difficult to draw conclusions about the value of the program relative to other obesity interventions because reported effectiveness measures and time frames vary. A cost-effectiveness analysis could also translate biometric effectiveness measures into a more global health metric such as the quality-adjusted life year (QALY), which represents a summary measure of quality of life that includes all positive and negative aspects of a health state [
19]. The use of QALYs would therefore allow cost-effectiveness analyses of obesity interventions to be directly compared to other childhood health interventions for obesity or other conditions.
Still, we can compare the cost of High Five to Kids to the cost of other pediatric obesity interventions. At $196 per child, the net cost of High Five for Kids falls within the range of four U.S. school- and community-based childhood obesity interventions, all calculated using a micro-costing approach and converted to 2011 dollars [
20]. Planet Health cost $18 per child, the Coordinated Approach to Child Health (CATCH) intervention cost $119 per child, the FitKid afterschool intervention cost $427 per child, and the STORY intervention had a cost ranging from $594-$994 [
5,
12‐
14]. Several analyses have evaluated the cost of obesity interventions outside of the U.S.; the cost of these interventions ranges from less than a dollar to $2,183 per child [
8,
16,
26,
27] when calculated using similar categories of costs and adjusted for inflation and purchasing power parity [
28].
Two phases of the Australian Live, Eat, and Play (LEAP) intervention, the only other primary care-based child obesity prevention program for which costs have been evaluated, were more costly than High Five for Kids. The net cost of delivering the LEAP intervention was approximately $852 (LEAP Phase 1) and $1,237 (LEAP Phase 2) per child from the health plan cost perspective. These estimates of the cost of LEAP include any reported costs of study-related visits with the primary care practitioner and equipment costs, but exclude any family-incurred time or out of pocket expenses [
9‐
11,
29]. Using similar categories of costs, the net cost of High Five for Kids ranged from $369-$725 per child. High Five for Kids was a more intensive intervention than LEAP, but LEAP was still more expensive than High Five for Kids, although some of the differences in costs may be attributable to differences in health care costs between countries and the use of nurse practitioners to conduct the intervention as opposed to physicians.
When making comparisons to other cost analyses, it is important to consider intervention characteristics or assumptions made in analyses that can affect per capita cost estimates. It is difficult to compare health plan perspective costs from other countries because the health services and unit costs may not be comparable [
30]. Moreover, analysis methods such as the choice of cost perspective (government, health plan, or societal), measurement of cost offsets due to changes in healthcare utilization, or the time frame of the analysis can make cost estimates from some studies appear more favorable than others. For example, a 2009 Assessing Cost Effectiveness (ACE) study by Moodie et al. models a theoretical primary care intervention based on LEAP, applied to a greater population (n = 9,485) than the initial LEAP1 study (n = 82) [
8]. A larger target population and the consideration of health care cost savings in adulthood from early weight loss results in a lower marginal cost per child for the ACE study than LEAP1. Given inconsistencies in methods and target populations, readers should be careful when interpreting and comparing the results of economic analyses, being sure to note the perspective, population size, and country of analysis, among other analysis methods and features of the intervention.
Our use of a micro-costing approach to calculate provider-incurred costs provides estimates of the marginal cost of the intervention per child treated, which may be useful to others planning programs. The micro-costing approach estimated substantially lower intervention costs than those generated using health plan perspective costs because health plan perspective costs may represent prices that could vary between practices and insurers, and may incorporate substantial administrative costs [
31]. The micro-costing approach, which considers average wages and unit costs, may better reflect resource utilization.
We excluded training costs from our primary analysis, consistent with the conventional framework for economic evaluation that these were start-up fixed costs that would not vary materially with the scale of the intervention [
32]. However, High Five for Kids APCs required repeated training sessions throughout the intervention time period that were not initially planned to refine their motivational counseling strategies. From that point of view, the costs of training APCs were not strictly start-up fixed costs and warranted examination in sensitivity analyses.
Because we were missing some health plan costs, we relied on imputation to derive expected costs from the health plan perspective. But because the children had a limited number of insurance providers and saw providers in one network, we believe that the imputed costs are fairly representative of actual costs. Secondly, the study was focused on a small number of practices in eastern Massachusetts, which may lead to concerns about the generalizability of study results to other practices in the U.S. However, the results found using the micro-costing approach represent national costs, improving generalizability of the findings. Lastly, the visit compliance among intervention subjects in primary care-based interventions such as High Five for Kids and LEAP was lower than expected and some usual care patients unexpectedly received weight counseling visits [
11]. As the majority of intervention costs were comprised of visit costs, our results may underestimate the cost of an intervention with higher compliance.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
EMT, MWG, SLG, LAP, and DRW were responsible for the study concept and design. EMT, MWG, KHH, and CMH were responsible for the acquisition of data. DRW was responsible for the analysis and interpretation of data and drafted the manuscript. All authors reviewed the manuscript for intellectual content and approved the final manuscript.