Download Full Text (336 KB)

Publication Date



Background: Obesity disproportionately affects children from low-income families, highlighting the need to improve treatment reach for this population. Family-based behavioral treatment (FBT) is an evidence-based multicomponent intervention to aid families in making pragmatic and impactful changes.

Objective: To assess the reach of FBT in two health systems serving low-income families.

Methods: FBT was implemented in rural Freeman Health System (FHS) and urban Children’s Mercy (CM) pediatric practices. Eligible patients included children ages 5-12 years with a BMI percentile ≥95th enrolled in Missouri Medicaid. The study was only offered in English. Patients were recruited through provider referrals during clinic visits or directly (letters/calls from intervention team). Patient reach was calculated by comparing the number and representativeness of referred and enrolled patients to all eligible patients and logistic regression was performed.

Results: 18.4% and 8.6% of eligible patients were referred to treatment in FHS and CM, respectively, with a higher proportion of providers in FHS (100%) making referrals than in CM (40.0%). Referral rates varied across providers, with FHS providers referring an average of 17.4% (standard deviation [SD]=8.3%) of eligible patients and CM providers referring an average of 10.9% (SD=24.2%) of eligible patients. Respectively at FHS and CM, 25.6% and 56.6% of referred patients enrolled in treatment, whereas only 5.8% and 9.5% of those recruited directly enrolled. Thus, referred patients were significantly more likely to enroll in treatment than patients recruited directly (FHS OR=4.15, 95% CI=2.15, 8.08, p<.001; CM OR=12.30, 95% CI=6.58, 23.58, p<.001). Latinx patients, in particular, were more likely to enroll via referral than via direct recruitment. 22% and 38% of patients enrolled via referral were Latinx, whereas 10% and 26% of patients enrolled via direct recruitment were Latinx, in FHS and CM respectively, though these differences were non-significant (FHS OR=2.50, 95% CI=.428, 14.6, p=0.281; CM OR=1.70, 95% CI=0.65, 5.571, p=0.357).

Conclusion: Although only a small proportion of eligible patients were referred to FBT by their primary care provider, provider referrals supported reach by increasing enrollment among patients. Encouragement from primary care providers may be valued highly by some patient families, and this may be especially true in certain population subgroups such as Latinx families. More work is needed to improve understanding of barriers to provider referrals to pediatric obesity interventions and to identify strategies for facilitating referrals that minimize burden on providers.

Document Type


Assessing Reach of Family-Based Behavioral Treatment for Pediatric Childhood Obesity Offered through Primary Care