CONSTRUCTION OF MORBIDITY TABLE FOR CRITICAL ILLNESS WITH REGIONAL CONSIDERATIONS

• Main Author: Magfirah Utami, Arli
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/80615
• Institution: Institut Teknologi Bandung

The morbidity table serves as a health analysis tool providing an overview of the severity and spread of diseases within a specific region. Considering regional aspects in constructing the morbidity table can offer more detailed insights into the impact of critical illnesses on public health in specific areas. This table incorporates age and the probability of an individual contracting a disease. Critical illnesses are defined as health conditions marked by vital organ dysfunction, high mortality risk without treatment, and potential for recovery. The selected critical illnesses, based on their frequency, include heart disease, stroke, and cancer. The process of constructing the table involves exposure calculation, claim calculation, morbidity rate calculation, and graduation. Data utilized in this table are sourced from membership and claim data from BPJS Kesehatan. Exposure calculation is derived from membership data, while claim calculation is obtained from claim data. Graduation employs the Natural Cubic Spline method due to its excellent smoothness, continuous curve, and adherence to the original data trend. The morbidity rates of each disease follow a similar pattern, resembling a bell curve with varying peaks. Heart disease exhibits a morbidity rate 5-6 times higher than stroke and 10-16 times higher than cancer. Regarding gender, heart disease shows no significant difference, with peaks for males and females occurring at ages 67 and 68, with probabilities of 0.0361057 and 0.0357369, respectively. Male stroke exhibits a higher probability compared to females, with peaks at ages 68 and 70, with probabilities of 0.0079331 and 0.0052521. Male cancer shows a lower probability compared to females, with peaks at ages 73 and 56, with probabilities of 0.0022158 and 0.0037764, respectively. More specific morbidity tables can be developed based on regions. Region division is obtained using the K-Means clustering method, then compared with the Regulation of the Ministry of Health of the Republic of Indonesia 2016. Division into 5 regional clusters reveals that clusters with high probabilities are clusters 2 (Bali, DKI Jakarta, South Sulawesi, South Sumatra, and North Sumatra) and 3 (parts of Sumatra, parts of Java, and South Kalimantan). Each region evidently influences the morbidity rate of a disease. The obtained results can inform the improvement of healthcare service quality and the formulation of insurance programs.