Molecular technologies have produced diverse arrays of animal models for studying genetic diseases and potential therapeutics. Many have neonatal phenotypes. Spinal muscular atrophy (SMA) is a neuromuscular disorder primarily affecting children, and is of great interest in translational medicine. The most widely used SMA mouse models require all phenotyping to be performed in neonates since they do not survive much past weaning. Pre-clinical studies in neonate mice can be hindered by toxicity and a lack of quality phenotyping assays, since many assays are invalid in pups or require subjective scoring with poor inter-rater variability. We find, however, that passive electrocardiography (ECG) recording in conscious 11-day old SMA mice provides sensitive outcome measures, detecting large differences in heart rate, cardiac conduction, and autonomic control resulting from disease. We find significant drug benefits upon treatment with G418, an aminoglycoside targeting the underlying protein deficiency, even in the absence of overt effects on growth and survival. These findings provide several quantitative physiological biomarkers for SMA preclinical studies, and will be of utility to diverse disease models featuring neonatal cardiac arrhythmias.