Respiratory infections among infants constitute a major burden to health care systems in developed nations, yet the course and risk factors leading to these conditions are poorly understood. We examine the longitudinal patterns of respiratory infection hospitalisation (RIH) and how these patterns are influenced by neonatal pulmonary morbidities. We included all live births (n = 429 058) occurring in the Australian state of Queensland between January 2009 and December 2015. Data were structured so that each participant had a record (present/absent) of RIH for each month from birth to 12 months. Initially, latent class growth analysis was used to identify the trajectories of RIH adjusted for spatial–temporal factors; using the identified trajectories of RIH as outcomes, we built a multinomial logistic regression model to identify neonatal predictors of RIH trajectories. Our results indicated that a four-class solution was the best fit to the data, comprising a ‘no-risk’ trajectory, a ‘low-risk’ trajectory, an ‘early-risk’ trajectory and a ‘chronic-risk’ trajectory. Compared with the no-risk trajectory, membership in the other trajectories was predicted by a range of neonatal pulmonary morbidities, with transient tachypnoea of newborn showing a specific relationship with the early-risk group and sleep apnoea showing a specific and strong risk with the chronic-risk group. Our findings suggest the possibility of identifying neonates at risk of recurrent RIH and implementing effective intervention strategies prior to neonatal discharge.