Abstract
Deprivation indices have been widely used to evaluate neighborhood socioeconomic status and therefore examine individuals within their regional context. Although some studies on the development of deprivation indices were conducted in Korea, additional research is needed to construct a more valid and reliable deprivation index. Therefore, a new deprivation index, named the K index, was constructed using principal component analysis. This index was compared with the Carstairs, Townsend and Choi indices. A possible association between infant death and deprivation was explored using the K index. The K index had a higher correlation with the infant mortality rate than did the other three indices. The regional deprivation quintiles were unequally distributed throughout the country. Despite the overall trend of gradually decreasing infant mortality rates, inequalities in infant deaths according to the deprivation quintiles persisted and widened. Despite its significance, the regional deprivation variable had a smaller effect on infant deaths than did individual variables. The K index functions as a deprivation index, and we may use this index to estimate the regional socioeconomic status in Korea. We found that inequalities in infant deaths according to the time trend persisted. To reduce the health inequalities among infants in Korea, regional deprivation should be considered.