Àá½Ã¸¸ ±â´Ù·Á ÁÖ¼¼¿ä. ·ÎµùÁßÀÔ´Ï´Ù.
KMID : 0614720140570050413
Journal of Korean Medical Association
2014 Volume.57 No. 5 p.413 ~ p.418
Use of big data for evidence-based healthcare
Ko Min-Jung

Lim Tae-Hwan
Abstract
Data pools and their integration are fueling the big data revolution in health care with the recent advances in information technology. Korea has shown tremendous promise in the utilization of big data for its advanced technology, computerized health data, and unique identifiers. However, the Personal Information Protection Act (PIPA) severely limits access to personal identifiers, which has discouraged the use of health data even for the public good. In contrast, western countries have focused on research without the use of identifiers, which has augmented the use of the available data while maintaining and respecting privacy; they have allowed some exemptions of informed consent and utilized limited data sets, which have the identifiers removed. The amount of research output has increased rapidly and an in-depth understanding of cancer has been made possible based on the linkage of Surveillance, Epidemiology, and End Results (SEER) and Medicare in the US. More than 700 projects covering a wide range of medical areas have been conducted, which has led to changes in clinical practice based on the Western Australian Data Linkage System. Although rare, evidence-driven decisions based on data linkage have been found in some cases in Korea; the adoption of prostate cancer screening as a national screening program was suspended as its cost-effectiveness has not been verified on the basis of data linkage by the National Evidence-Based Healthcare Collaboration Agency. For the active use of health data, there is an urgent need to amend PIPA, prepare regulations for data analysis, and foster collaboration among data-related institutions. Great projects based on data linkage will guarantee the world¡¯s leading research output and will be major sources for moving forward to success.
KEYWORD
Big data, Data collection, Evidence-based practice
FullTexts / Linksout information
 
Listed journal information
ÇмúÁøÈïÀç´Ü(KCI) KoreaMed