• Health Policy and Management
• /
• v.20 no.3
• /
• pp.21-35
• /
• 2010

This study examined the changes in the service mix of Korean hospitals in 6 metropolitan cities between 2003 and 2005, and assessed whether the sample hospitals exhibit consistent trend or chance variation in multiple years. Three measures of hospital service mix, focusing on the specialization of services, were applied: information theory index, internal Herfindahl index, and number of distinct diagnosis-related groups (DRGs) treated. National Health Insurance claims were used to calculate the indexes. Specialization indexes were calculated in each year, and then examined to identify the pattern over time. Kappa analysis was applied to assess the agreements of specialization score between two years after hospitals were categorized into 4 groups with quartiles. Kappa score showed that the service mix of hospitals were changed during the study years. Specialization scores were increased given the market structure for three years. Hospitals which showed higher or lower specialization scores than the average of the scores consistently classified into the same group. Specialization indexes showed relatively consistent pattern over 3 years, and such consistencies were evident for hospitals regardless of the specialization status. Policy makers can identify the degree of specialization with the indexes, and it could provide a picture of how hospital services were mixed and changed over time.

• Health Policy and Management
• /
• v.18 no.2
• /
• pp.67-85
• /
• 2008

This study evaluated the specialization status of Korean hospitals by applying index measures that were developed to determine how hospitals are specialized. In addition, multivariate regression analysis was applied to assess how the measures responded to the internal and external factors of hospitals. National Health Insurance claims for 2004 were used to calculate the information theory index, internal Herfindahl index, number of distinct diagnosis-related groups (DRGs) treated, and percent of the five most common DRGs. Data from the Ministry of Health and Welfare and Korean Hospital Association were used to determine the size, ownership, teaching status, organization type, and location of the hospitals. The four indexes analyzed showed that there were significant differences in the specialization status of providers, depending on the provider size, organization type, and location. Hospitals that were smaller and located in metropolitan areas tended to provide specialized services; this is considered to constitute a competitive strategy for hospitals. It is expected that specialized hospitals will increase given the current market structure. Therefore, policy makers will need an index for measuring how hospital services are specialized. Information from such an index could provide a picture of how hospital services are mixed and change over time.

• Health Policy and Management
• /
• v.26 no.2
• /
• pp.107-114
• /
• 2016

Background: Over the last few decades, because hospitals in South Korea also have undergone dramatic changes, Korean hospitals traditionally have provided specialized health care services in the health care market. Inner Herfindahl-Hirschman Index (IHI) measures hospital caseloads based on patient proportions, independent of patient volumes. However, IHI that rely solely on patient proportions might be problematic for larger hospitals that provide a high number of diagnosis categories, as the patient proportions in each category are naturally relatively smaller in such hospitals. Therefore, recently developed novel measure, category medical specialization (CMS) is based on patient volumes as well as patient proportions. Methods: We examine the distribution of hospital specialization score by hospital size and investigate association between each hospital specialization and length of stay per case and hospital cost per case using Korean National Health Insurance Service-cohort sample data from 2002 to 2013. Results: Our results show that IHI show a decreasing trend according to the number of beds and hospital type but CMS show an increasing trend according to the number of beds and hospital type. Further, inpatients admitted at hospitals with higher IHI and CMS had a shorter length of stay per case (IHI: B=-0.104, p<0.0001; CMS: B=-0.044, p=0.001) and inpatients admitted at hospitals with higher IHI and CMS had a shorter hospital cost per case (IHI: B=-0.110, p=0.002; CMS: B=-0.118, p=<0.0001). Conclusion: This study may help hospital policymakers and hospital administrators to understand the effects of hospital specialization strategy on hospital performance under recent changes in the Korean health care environment.