A prospective case control study to determine the association of early introduction of solids with admission to hospital with pneumonia was undertaken at Mount Hagen General Hospital (MHGH) in the highlands of Papua New Guinea (PNG) over a 3-month period in 2005. Twenty three infants up to 6 months of age admitted with radiologically confirmed pneumonia were compared with 24 infants of similar age attending the well baby clinic for immunization and with 35 infants admitted to the hospital with conditions other than pneumonia or meningitis. There was a highly significant difference in feeding patterns between the groups. Children with pneumonia were much more likely than the control children to have started solids before the age of 2 months [OR=18.06 (4.8–72.86)]. They were also significantly more likely to have been admitted previously with a diagnosis of pneumonia (P < 0.001). The children in each group were of comparable age and weight and there were no obvious confounding factors. This study provides clear evidence for the association between early introduction of solids and pneumonia in PNG highlands children. The findings are consistent with other international data. While the reasons for the association remain speculative, the association strongly reinforces the need to educate the community on best infant feeding practices and to discourage the early introduction of solids.