A Network Meta-Analysis About Effectiveness of Mind-Body Therapies in Pediatric Palliative Care

To evaluate and compare the effectiveness of mind-body therapies for managing symptoms and improving quality of life in children receiving palliative care using a network meta-analysis (NMA).

A systematic review and network meta-analysis (PRISMA-NMA guidelines) registered with PROSPERO (CRD42020198399).

Comprehensive searches were conducted across seven databases and trial registries up to June 2024. Eligible studies involved randomized and nonrandomized trials with children (0-18 years) with life-limiting conditions receiving mind-body therapies. Primary outcomes were symptom control (pain, anxiety, emotional distress) and quality of life indicators. A frequentist NMA was performed using the netmeta package in R. Transitivity was assessed by comparing clinical and methodological characteristics across studies.

Forty-three studies (31 randomized trials), primarily in pediatric oncology, were included. Statistical heterogeneity was high (I² > 90%). Most network meta-analysis comparisons were inconclusive due to wide confidence intervals. For pain reduction, indirect hypnosis (SMD -0.91, [95% CI -1.78 to -0.03]) showed point estimates favoring the intervention, but confidence intervals were wide. Notably, most evidence for pain derived from procedural pain studies in pediatric oncology; these findings should not be generalized to chronic disease-related pain or nononcologic palliative populations without caution. For anxiety reduction, indirect hypnosis showed a statistically significant benefit (SMD -2.40, [95% CI -4.73 to -0.06]), while direct hypnosis suggested a potential but imprecise benefit. Comparisons for other therapies (music therapy, mindfulness, aromatherapy) were largely inconclusive, with most confidence intervals crossing the null. P-score rankings suggested a hierarchy based on imprecise point estimates; hypnosis-based interventions had the highest probability of being the best therapy. However, these rankings should be interpreted as hypothesis-generating only.

This first NMA in pediatric palliative care suggests that hypnosis-based interventions may be promising for symptom management, but the evidence is limited by high heterogeneity, imprecision and a predominance of procedural pain studies in oncology. These findings generate hypotheses for future research and highlight an urgent need for more rigorous, head-to-head comparative trials to establish a robust evidence base for integrative care in this vulnerable population.