Multimodal health behaviour-changing interventions for children living with obesity and their parents

Childhood obesity is a global public health problem with multiple causes. First-line obesity interventions target changes in health-related behaviours (diet, physical activity, behaviour). However, the sustainability and long-term results of these interventions are uncertain.

To assess the effects of multimodal health behaviour-changing interventions for children under 10 years living with obesity and their parents.

We used CENTRAL, MEDLINE, three other databases, and two trial registers, together with reference checking and contact with study authors, to identify studies included in the review. The latest search date was 28 February 2024.

We included randomised controlled trials in children under 10 with obesity (and their parents) that tested multimodal health behaviour interventions (diet, physical activity, and/or behaviour change) versus control (no treatment, usual care, or waiting list), with at least one year of follow-up.

Critical outcomes were physical well-being, mental well-being, physical activity, health-related quality of life (HRQoL), obesity-associated disability, adverse events, and anthropometry (body mass index (BMI) z-score).

We used the original version of the Cochrane Collaboration's tool for assessing risk of bias (RoB 1).

We synthesised outcomes using random-effects meta-analysis where possible, otherwise narratively. For continuous outcomes on the same scale, we calculated mean differences (MDs) with 95% confidence intervals (CIs). We calculated standardised mean differences (SMDs) with 95% CIs when studies used different instruments for the same outcome or when applying a generic minimally important difference. We assessed certainty of evidence for critical outcomes with GRADE.

We included 34 RCTs involving 6849 participants, aged four to nine, conducted in high-income countries. We identified 23 healthcare-based interventions implemented in primary care and hospitals, and 11 community-based interventions implemented in schools and community centres. The intervention components included sessions on dietary modification, physical activity, and behavioural change. Three studies targeted parents directly, and the remainder used family-based approaches, primarily through individual or group sessions.

Healthcare-based multimodal interventions versus control Healthcare-based multimodal interventions likely result in little to no difference in physical well-being at 12 months of follow-up (Pediatric Quality of Life Inventory (PedsQL) physical score, final value: SMD 0.05, 95% CI -0.09 to 0.19; I² = 0%; 2 studies, 757 participants; moderate-certainty evidence). When measured on the PedsQL physical functioning or DUX-25 physical subscale as a change-from-baseline score, these interventions may improve physical well-being at 12 months, but the evidence is very uncertain (SMD 1.16, 95% CI -0.73 to 3.04; I² = 85%; 2 studies, 56 participants; very low-certainty evidence). They likely result in little to no difference in mental well-being at 12 months of follow-up (PedsQL emotional or psychosocial scores; SMD 0.02, 95% CI -0.12 to 0.16; I² = 0%; 2 studies, 757 participants; moderate-certainty evidence). These interventions may improve objectively-assessed physical activity at 12 months (SMD 0.23, 95% CI -0.16 to 0.63; I² = 60%; 3 studies, 278 participants; low-certainty evidence), but may result in little to no difference in subjectively-assessed physical activity at 12 months (SMD 0.01, 95% CI -0.20 to 0.22; I² = 47%; 5 studies, 776 participants; low-certainty evidence) and 24 months (SMD 0.15, 95% CI -0.47 to 0.77; I² = 82%; 2 studies, 275 participants; low-certainty evidence). They likely result in little to no difference in HRQoL at 12 months (MD 0.36, 95% CI -1.18 to 1.90; I² = 0%; 5 studies, 921 participants; moderate-certainty evidence). We are very uncertain about the effects of these interventions on adverse events (3 studies, 614 participants; very low-certainty evidence). The evidence is very uncertain about the effect of healthcare-based multimodal interventions on BMI z-score at 12 months (MD -0.15, 95% CI -0.23 to -0.06; I² = 68%; 16 studies, 2397 participants; very low-certainty evidence). They may result in little to no difference in BMI z-score at 24 months (MD -0.08, 95% CI -0.18 to 0.02; I² = 65%; 4 studies, 608 participants; low-certainty evidence). None of the studies reported obesity-associated disability. Community-based multimodal interventions versus control Community-based multimodal interventions may result in little to no difference in physical well-being (MD 2.92, 95% CI -3.63 to 9.47; I² not applicable; 1 study, 87 participants; low-certainty evidence) and likely result in little to no difference in mental well-being at 12 months of follow-up (SMD -0.04, 95% CI -0.23 to 0.16; I² = 0%; 2 studies, 428 participants; moderate-certainty evidence). These interventions may result in little to no difference in objectively-assessed physical activity at 12 months (SMD -0.02, 95% CI -0.27 to 0.23; I² = 39%; 4 studies, 481 participants; low-certainty evidence) and likely result in little to no difference at 24 months (SMD 0.06, 95% CI -0.15 to 0.26; I² = 0%; 2 studies, 362 participants; moderate-certainty evidence). These interventions may result in little to no difference in HRQoL (final score) at 12 months (MD 0.58, 95% CI -2.11 to 3.27; I² = 19%; 4 studies, 653 participants; low-certainty evidence), but may lead to a small benefit in HRQoL (change-from-baseline score) at 24 months (MD 4.30, 95% CI -0.76 to 9.36; I² not applicable; 1 study, 121 participants; low-certainty evidence). They likely result in little to no difference in BMI z-score at 12 months (MD 0.02, 95% CI -0.08 to 0.12; I² = 63%; 5 studies, 420 participants; moderate-certainty evidence) and 24 months (MD 0.01, 95% CI -0.11 to 0.12; I² = 0%; 2 studies, 190 participants; moderate-certainty evidence). None of the studies reported adverse events or obesity-associated disability.

Multimodal health behaviour-changing interventions may improve objectively-assessed physical activity at 12 months when delivered in healthcare settings and may slightly improve HRQoL at 24 months when delivered in the community, but they may have little to no effect on the other pre-defined critical outcomes, including physical and mental well-being, and anthropometry (BMI z-score). Future research should explore innovative approaches to the care of children living with obesity and ensure the inclusion of diverse populations, given the limited evidence from disadvantaged or culturally/ethnically diverse groups, and from low-resource settings.

World Health Organization (WHO) REGISTRATION: Protocol (2024): PROSPERO CRD42023468867.