Abstract
The objective of this study was to examine the feasibility of an outpatient casemix classification system based on the health insurance database in Thailand. The Ambulatory Patient Groups (APG) developed by 3M and Adjusted Clinical Group (ACG) by John Hopkins University were used as the reference model. Model for predicting outpatient expenditure in selected chronic disease groups i.e. diabetes mellitus, hypertension and high cost of care i.e. renal failure and thallasaemia. From 39.1 outpatient visits was grouped into 82 medical APGs which is 21.2 million visits. Most common medical APG groups were 3.3 million visits in influenza, upper respiratory infection and ear, nose throat infections (APG 542), 1.5 million visits in dental diseases (APG 541) and 1.4 million visits in hypertension (APG 572). While can be grouped into 80 groups in ACG which is 7.7 million persons. Most common ACG groups were 1.9 million persons in acute minor, age > 5 (ACG 300), 8.7 million persons in chronic medical: stable (ACG 900) and 0.7 million persons in likely to recur without allergy (ACG 500). From APG concept, significant variable (p-value < 0.05) which increasing medical charges per visit in diabetes model were age, male, provincial and general hospital, refer-out, complication and comorbidities as hypertension, cardiovascular diseases (CVD), cerebrovascular diseases and renal diseases. Significant variable which increasing medical charges in hypertension model were age, male, provincial and general hospital and co-morbidities as diabetes, hyperlipidaemia, cerebrovascular diseases, ischaemic heart disease and renal diseases. Significant variable which increasing medical charges renal failure model were provincial and general hospital, APG and cardiovascular diseases as comorbidity. Age was significant variable which increasing medical charges in thallasaemia model. From ACG concept, significant variable (p-value < 0.05) which increasing medical charges in chronic diseases with hypertension were age, male, provincial and general hospital, number of visits, length of duration between visit and MAC 5 AND 15. Significant variable which increasing medical charges in diabetes model were age, provincial and general hospital, number of visits, length of duration between visit. The results indicated that outpatient classification system requiring more information such as significant procedure and ancillary service for the other two main APG groups. These will be further developed for outpatient classification system in Thailand.
