Abstract
Mental health service is an important health service system with increasing resource consumption due to social and economic changes. The objective of this study was to develop Thai Mental Health Casemix Classification (TMHCC) for inpatient payment method. This research and development employed 4 steps to develop TMHCC: 1) standard data identification, 2) data collection and edit, 3) cost calculation 4) classification, calibration and evaluation. Eighteen hospitals with inpatient beds for psychiatric care and 2015 top-down cost accounting data joined the study organized by the Department of Mental Health. Data for TMHCC development included administrative data, clinical data, resource uses and mental problems. The Thai version of the Health of the Nation Outcome Scales (HoNOS) and the Health of the Nation Outcomes Scales for Child and Adolescent mental health (Thai-HoNOSCA) were used for adult and children psychiatric assessment. Prospective data collection was conducted in the second half of fiscal year 2016. The total inpatient data for classification were 15,417 cases. The statistics to explain performance of the TMHCC were coefficient of variation (CV) for homogeneity of cases within each group and the reduction in variance (RIV) for overall performance of the casemix classification. Results: TMHCC classification provided 39 disease groups (34 for adult and 5 for children). The highest number of adult cases was for Schizophrenia, Age <= 50, with cognitive problems, with problems from reactive/aggressive/ agitated behavior (2,088 cases, 17.1% of all cases). About 83.4% of children patients were grouped into Other Child Psychiatric Conditions (2,463 cases). Mental Health Disease, Ongoing type had the highest cost per admission (200,817 Baht) and Mental Health Disease, Same-day type had the lowest cost per admission (590 Baht). The result of CV on cost showed that only one group had CV higher than 100%. Most TMHCCs had high homogeneities within groups. The results also confirmed good overall performance of TMHCC that RIV on cost after the classification was as high as 39.9%. Moreover, RW of TMHCC better predicted cost than RW of the current Thai DRG version 5. In summary, the performance of this casemix is good enough based on homogeneity within groups and high heterogeneity of resource uses between groups. Recommendations: The payment based on TMHCC was an option for payment. However, payers should prepare budget for psychiatric inpatient care benefit package that needed longer stay service in hospital than acute inpatient. Moreover, mental health measurement data should be put into addition¬al routine electronic databases of all hospitals that provided care to psychiatric inpatients.
