Study Overview
Objective. To determine the efficacy, both short and long term, of a behavioral intervention targeting overweight parents and their children simultaneously versus an intervention focused on weight management only for the child within the patient-centered medical home (PCMH).
Design. 4-center, 2-arm, randomized controlled trial.
Setting and participants . Study participants were recruited from 4 urban/suburban pediatric practices. Primary care providers (PCPs) recruited patients at the time of well or sick visits based on body mass index (BMI) flagged prior to the visit by Patient Enhancement Assistants (PEAs). 171 parent/child dyads were assessed for eligibility and 105 were randomized in blocks of 12 dyads using a random number generator and stratified by child’s gender. Pediatricians were blind to their patient’s group assignments. Inclusion criteria were as follows: children aged 2–5 with a BMI higher than the 85th percentile for both age and gender, and 1 parent with a BMI greater than 25. Exclusion criteria were limited to children who were small for gestational age and/or short stature, and child or parent inability to perform physical activity. Specific precautions were taken to prevent contamination between intervention and information control (IC) groups [1].
Intervention. Three PEAs who held a masters or bachelors degree in psychology, nutrition, exercise science, or equivalent, or were registered dietitians, were embedded within each PCMH practice. For both the intervention and IC groups, parents attended 13 one-hour group sessions led by a PEA over a 12-month period, followed by a 12-month follow-up period with 3 additional meetings. The PEA telephoned parents between scheduled meetings. Pediatricians reviewed child’s weight changes every 6 months during scheduled appointments and the PEA sent progress notes in between these visits [2]. Dietary, physical, and sedentary activity guidelines were given based on the recommendations of a national multi-organizational expert committee [3]. Parents were given specific goals for their child, including a 0.5- to 1-pound per week loss, 60 minutes per day of physical activity, and limiting TV and screen time to less than 2 hours per day.
In addition, the intervention group received parenting and behavior change strategies to promote both parent and child weight loss. Parents were instructed to weigh themselves and their child once per week and monitor physical activity and diet. They received individual meetings with the PEA before or after group meetings to review goal setting and food/physical activity diaries. Parents were also given a weight loss goal of 1 to 2 pounds per week and were advised to model physical activity by engaging in active play with their child for at least 10 minutes per day.
Main outcome measures. The main outcome measures were %0BMI and z-BMI. Percent 0BMI is defined as [(child’s BMI – 50th percentile BMI)/50th percentile BMI] x 100 [2]. The authors chose %0BMI as the primary outcome measure because z-BMI can diminish the effect of weight change in heavier children [4]. Both measures were expressed as mean ± standard error (SEM). Parent weight change was measured using BMI alone.
The child’s weight was measured at each session and height was measured at baseline, 3, 6, 12, 18, and 24 months. Parent weight was measured every session in the intervention group, but only at baseline, 6, 12, 18, and 24 months in the IC group. A standardized protocol was followed for all height and weight measurements. An intention to treat analysis (ITT) was conducted on all parent/child dyads, regardless of whether or not they completed the study ( n = 96).
Results. Research assistants assessed 171 parent/child dyads for eligibility. 66 were excluded for either not meeting inclusion criteria ( n = 24) or declining to participate ( n = 42). 105 dyads were randomized, but 9 did not receive the allocated intervention because they did not start the study, resulting in a total of 96 dyads included in analysis: 46 in the intervention group and 50 in the IC. Twelve- and 24-month completion rates were 83% and 73% respectively; there was no difference in attrition between intervention and IC groups.
The mean child ages of the intervention and IC groups were 4.6 ± 0.2 and 4.4 ± 0.2 years, respectively. 33 of the 46 children in the intervention group and 37 of the 50 children in the IC group were identified as non-Hispanic white. The mean yearly income of all families was $65,729 ± $3068, with only 8.3% of families below $20,000.
The intervention group had greater decreases in child %0BMI from baseline to 6, 12, 18, and 24 months than the IC group. Similar trends were seen with child z-BMI. A slower increase in height was observed in the intervention group when compared with the IC at both 18 months ( P < 0.001) and at 24 months ( P < 0.02). Parents showed greater overall BMI reduction in the intervention group as opposed to the IC group at all time points ( P < 0.001). BMI changes achieved at 6 months were maintained at 24 months. %0BMI and parent BMI changes were correlated from baseline to 12, 18, and 24 months. No significant baseline moderators were found among the children in either group.
Conclusion. This study demonstrated that within the PCMH model of pediatric primary care, an intervention focused on joint behavior change and weight modification treatment of parents and children led to better initial and sustained improvements in %0BMI and z-BMI (in children) and BMI (in parents) than a child-focused IC.
Commentary
Over one-third of children and adolescents are considered to be overweight or have obesity, a number that has doubled in the past 30 years [5]. Pediatrician and primary care physician visits are optimal places to identify overweight children who are at risk for obesity and begin prevention measures, although identifying overweight and obese younger children can be difficult [6]. This study used PEAs to aid physicians in identification, implementation, and delivery. With increasing evidence to support pediatrician involvement in intensive weight management in a primary versus specialty care setting, embedding PEAs within the PCMH model may be an important way to help deliver care for overweight/obese children [7].
Although many approaches have been considered to target childhood obesity, this study represents an important contribution to the literature because it demonstrates that a primary care–based intervention targeting parents as well as their young children is more efficacious for weight management than a more traditional, child-only focused intervention. In addition, the intervention included many different evidence-based components such as teaching behavior modification techniques to parents, consideration of parenting styles and techniques, and encouraging simultaneous parental weight modification. While the U.S. Preventive Services Task Force (USPSTF) recommends intensive interventions with 30 sessions over 2 years [8], this study was able to accomplish significant weight change in 13 sessions.
This intervention is unique in its integration of parenting techniques with other evidence-based strategies for child weight management. Although it has been shown in the literature that certain parenting styles can positively impact children’s health behaviors [9], namely the use of positive reinforcement and monitoring children’s health practices [10], only a few studies have looked at the impact of parenting interventions on childhood obesity. Mazzeo et al demonstrated a significant reduction in child BMI with a parenting-only intervention in the NOURISH trial [11], Slusser et al found a significant child BMI reduction using parent training for low-income, 2- to 4-year-old children [12], and Brotman et al conducted a longitudinal study demonstrating that a family intervention could decrease BMI and improve overall child health behaviors [13]. Despite these aforementioned studies, there is a lack of longitudinal data on the association between general parenting style and weight [14], and this study addresses this gap in literature by providing 2-year follow-up and demonstrating sustained impact on the intervention group.
This study had many additional strengths, including randomized design, primary care physician blinding, use of intention to treat analysis, standardization of measurement tools, clear justification of sample size, long-term follow-up, and the use of child-appropriate BMI measures (eg, %0BMI vs. z-BMI as primary outcome measure). In addition, the intervention setting in a PCMH follows the trend of increasing interest in exploring this model of health care delivery [15,16]. It is also important to note that the intervention and IC groups received the same number of group visits and phone calls, the only difference being the content and the extra 1:1 PEA sessions received by the intervention group.
The few weaknesses include that the PEAs could not be blinded to treatment allocation, and generalizability is limited by the mostly non-Hispanic white population and that only 8.3% of the study population had an annual household income of less than $20,000. All parents included in this study were on the high end of the obese range (BMI 30–39.9), with baseline BMI values of 37.2 and 36.2 in the intervention and IC groups respectively. In addition, the age of the children included in the study were on the high end of the designated 2- to 5-year-old range: 4.6 years (IC) and 4.4 years (intervention). Although findings were promising within this specific population, further research in younger and more diverse populations is necessary [11].
Finally, it is unclear whether this intervention is scalable, and a cost-effectiveness analysis of this intervention is needed. This study was designed to limit the PCP’s role and simplify the process of identifying and intervening on overweight children and their parents, yet this required 3 part-time PEAs and a project coordinator responsible for delivering all of the group sessions and providing follow-up counseling to both intervention and IC groups.
Applications for Clinical Practice
This study demonstrates that in a mostly white, urban/suburban population, a parenting and behavior modification intervention focused on both parent and child leads to greater improvements in %0BMI and z-BMI in the child and BMI reduction in parents compared with an intervention focused on the child alone within pediatric PCMH practices. This intervention should be tested in more diverse populations. This study also suggests further exploration of the use of PEAs to help clinicians address obesity within the PCMH model of primary care.
—Natalie Berner, BA, and Melanie Jay, MD, MS