In the Thai Universal Coverage scheme, hospital providers are paid for their inpatient care using Diagnosis Related Group (DRG) reimbursement. Questionable quality of the submitted DRG codes has been of concern whereas knowledge about hospital coding practice has been lacking. The objectives of this thesis are (1) To explore hospital coding practice and its determinants, (2) To identify the ways hospitals may use to manipulate the DRG system, and (3) To describe and compare the characteristics of hospitals that were included in the Summary and Coding Audit Database (SCAD) and those that were not. These three objectives are answered by three types of analysis: a case study of hospital coding practice, systematic literature review, and a nationwide hospital survey. In the case study, interviews with responsible staff in selected hospitals suggested that hospital coding practice comprises structure and process components. The structural component includes human resources, hospital committee, and information technology infrastructure. At least eight health care professional disciplines are involved in the coding process which comprises seven major steps: (1) Discharge Summarization, (2) Completeness Checking, (3) Diagnosis and Procedure Coding, (4) Code Checking, (5) Relative Weight Challenging, (6) Coding Report, and (7) Internal Audit. Hospital coding practice can be affected by five factors: (1) Internal Dynamics, (2) Management Context, (3) Financial Dependency, (4) Resource and Capacity, and (5) External Factors. The literature review, which was conducted using seven databases clarified that DRG creep (defined as a deliberate and systematic shift in a hospital's reported case-mix in order to improve reimbursement) is not merely about upcoding. Indeed, DRG system can be manipulated by hospitals using three different approaches I have called "3C": Corporate, Clinical, and Coding Practices, which are, respectively, the responsibility of hospital management, the clinical care team, and hospital coders. The questionnaire developed from case study findings was used to survey 920 hospital included in SCAD and 390 Non-SCAD hospitals. The overall response rate was 37.56%. Characteristics and hospital coding practices of the two groups were found to be different. Exploratory Factor Analysis was used to simplify 63 opinion-based questionnaire items to a 3-factor model with 20 items, which can be used to develop a Hospital Coding Practice Scale. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]