Access to health care is complicated to define. It is a multidimensional process. In addition to the matters of quality of care, geographical accessibility and availability of the right type of care, finance, and acceptability are all involved. The purposes of this paper are to measure the geographic distances between patient residency locations and health service organizations in which the patients hadvisited, and to investigate the association between geographical distance measures and variables involved in health service utilization. The study used the first and the second wave of the 2008 Korea Health Panel Survey. The samples of analyses were patients who had visited outpatient or used ambulatory health services, and the total observations (visit numbers) analyzed were 229,128. We divided the samples into a frequent-visit illness group (Group 1) and a non-frequent visit illness group (Group 2) based on over 5,000 total visit numbers. We exploited three level analyses using xtmixed of STATA${Box}$ 11.1 command with/without interaction terms among age, sex, and occupation. Geographical distances were measured using the Haversine method. Group 1 was tended to older and lower equivalent income than those of Group 2, but the geographic difference were not observed in terms of area deprivation index and standard mortality ratios. Amongst group 1, diabetes mellitus patients travelled far to visit health care organizations, and arthritis patients were more deprived in terms of the personal and areal characteristics. The study revealed that residents in rural areas traveled about 10 times more long distances than those whom lived in larger cities after adjusting for various variables, which we used for analyses. This study contributed to the practical understanding of health service utilizations using empirical analyses, and found that the types of diseases and socioeconomic characteristics of patients tended to define the amount of travel distance to healthcare organizations.