To assess whether underlying diagnosis affects morbidity and mortality outcomes in patients with chronic respiratory failure, we studied 55 patients with chronic respiratory failure of infancy and childhood (CRFIC). Entry criteria included patients with chronic respiratory failure due to static neurologic or neuromuscular conditions or secondary to other disease processes considered likely to improve or resolve over time. Subjects were grouped into those having chronic lung disease (CLD, n = 22), neurologic or neuromuscular diseases (NM, n = 21), or congenital abnormalities affecting the respiratory system (CA, n = 12). The average duration of follow-up was 21.3 months. There were no differences between groups in mortality with only four deaths (7%). Patients with CLD fared better than those with NM or CA in duration of ventilatory support, duration of tracheostomy, percentage of successful weaning from mechanical ventilation, and neurodevelopmental outcomes. Subjects with CLD had a significantly greater frequency of tracheomalacia (86%), feeding disorders (86%), and hypogammaglobulinemia G (77%). There were no differences between groups for respiratory readmissions or family dysfunction. We conclude that almost all patients with CRFIC will survive, but morbidity outcomes will vary based on the underlying diagnosis.