Short-term causal effects of common treatments in ambulatory children and young adults with cerebral palsy: three machine learning estimates.

Orthopedic and neurological impairments (e.g., muscle contractures, spasticity) are often treated in children and young adults with cerebral palsy (CP). Due to challenges arising from combinatorics, research funding priorities, and medical practicalities, and despite extensive study, the evidence base is weak. Our goal was to estimate the short-term effectiveness of 13 common orthopedic and neurological treatments at four different levels of outcome in children and young adults diagnosed with CP. The outcome levels considered were body structures, specific gait kinematic deviations, overall gait kinematic deviations, and functional mobility. We used three well-establish causal inference approaches (direct matching, virtual twins, and Bayesian causal forests) and a large clinical gait analysis database to estimate the average treatment effect on the treated (ATT). We then examined the effectiveness across treatments, methods, and outcome levels. The dataset consisted of 2851 limbs from 933 individuals (some individuals underwent multiple treatment episodes). Current treatments have medium effects on body structures, but modest to minimal effects on gait and functional mobility. The median ATT of 13 common treatments in children and young adults with CP, measured as Cohen's D, bordered on medium at the body structures level (median [IQR] = 0.42 [0.05, 0.60]) and became smaller as we moved along the causal chain through specific kinematic deviations (0.21 [0.01, 0.33]), overall kinematic deviations (0.09 [0.03, 0.19]), and functional mobility (-0.01 [-0.06, 0.13]). Further work is needed to understand the source of heterogeneous treatment effects, which are large in this patient population. Replication or refutation of these findings by other centers will be valuable to establish the generalizability of these results and for benchmarking of best practices.