ANALISIS UTILIZATION RISK DAN PEMETAAN RISIKO PEMBIAYAAN ISPA NON-SPESIALISTIK DI FKRTL MENGGUNAKAN TWO-PART COLLECTIVE RISK MODEL BERBASIS REGRESI LOGISTIK DAN GAMMA GLM

Alfia Sabrina Oktavia, Danang Teguh Qoyyimi, M.Sc., Ph.D.

2026 | Skripsi | S1 ILMU AKTUARIA

Pergeseran penanganan kasus Infeksi Saluran Pernapasan Akut (ISPA) non spesialistik dari Fasilitas Kesehatan Tingkat Pertama (FKTP) ke Fasilitas Kesehatan Rujukan Tingkat Lanjutan (FKRTL) memicu avoidable cost yang membebani Dana Jaminan Kesehatan Nasional (JKN). Penelitian ini bertujuan mengkuantifikasi avoidable cost dan memetakan risiko finansial tersebut antar-wilayah menggunakan pendekatan two-part Collective Risk Model pada Data Sampel Klaim BPJS Kesehatan periode 2015–2024. Kasus avoidable diidentifikasi melalui pendekatan rule based berdasarkan kriteria ICD-10 dan dievaluasi menggunakan Cost-Minimization Analysis. Komponen frekuensi rujukan dimodelkan menggunakan regresi logistik biner, sedangkan komponen severity biaya dimodelkan menggunakan Gamma Generalized Linear Model. Expected loss hasil pemodelan individu kemudian diagregasikan ke tingkat provinsi untuk dipetakan ke dalam policy matrix dua dimensi berbasis perspektif absolut dan intensitas. Hasil analisis menunjukkan bahwa mayoritas kasus ISPA di tingkat rujukan tergolong sebagai kasus non-spesialistik, dengan karakteristik struktur risiko yang didominasi oleh tingginya komponen frekuensi rujukan dibandingkan besaran biaya pelayanan. Melalui policy matrix versi intensitas, penelitian ini berhasil menangkap heterogenitas risiko antar-provinsi secara spasial, membedakan wilayah dengan kurang optimalnya sistem rujukan kronis seperti Su matera Utara, serta memisahkan wilayah yang beban pembiayaannya murni dipicu oleh skala populasi seperti Jawa Barat. Penelitian ini berkontribusi menyediakan ke rangka analisis risiko aktuaria terintegrasi guna mendukung pengelolaan utilization risk dan keberlanjutan Dana JKN berbasis wilayah.

The shift of non-specialist Acute Respiratory Infection (ARI) cases from primary health care to secondary referral facilities generates avoidable costs that burden Indonesia’s National Health Insurance Fund. This study aims to quantify the avoidable cost and map regional financial risks using a two-part Collective Risk Model framework on BPJS Kesehatan Claims Sample Data for the 2015–2024 period. Non-specialist ARI cases were identified via a rule-based ICD-10 approach and evaluated using Cost-Minimization Analysis. Within the two-part CRM, the frequency component was modeled using binary logistic regression, while the severity component utilized a Gamma Generalized Linear Model. Individual expected losses were aggregated to the provincial level for mapping via a two-dimensional policy matrix from both absolute and intensity perspectives. Results reveal that the majority of ARI cases at referral facilities were classified as non-specialist, with the over all risk structure heavily dominated by referral frequency rather than service severity. Through the intensity policy matrix, this study successfully captured regional risk variabilities, distinguishing provinces with chronic sub-optimality structural referral, such as North Sumatra, from those whose financial burdens are primarily driven by population scale, such as West Java. This study contributes an integrated actuarial risk analysis framework to support regional utilization risk management and enhance the long-term sustainability of Dana JKN.

Kata Kunci : avoidable cost; Collective Risk Model; expected loss; Jaminan Kesehatan Nasional; utilization risk.