Diabetes Melitus (DM) membutuhkan tata laksana pengobatan kronis jangka panjang yang dikelola melalui Program Rujuk Balik (PRB) di Fasilitas Kesehatan Tingkat Pertama (FKTP) untuk meminimalkan beban pelayanan di rumah sakit. Namun, capaian pemanfaatan insulin PRB di fasilitas kesehatan tingkat primer belum optimal akibat berbagai determinan dari sisi pasien (demand side) maupun fasilitas kesehatan (supply side). Tujuan penelitian ini adalah diketahuinya hubungan faktor determinan tingkat individu dan fasilitas kesehatan terhadap proporsi pemanfaatan insulin PRB. Penelitian ini merupakan studi analitik observasional dengan pendekatan potong lintang (cross-sectional) menggunakan data sekunder dari aplikasi Self-Service Business Intelligence (SSBI) BPJS Kesehatan di lima Kabupaten/Kota Provinsi Jambi periode pelayanan tahun 2024. Total sampel sebanyak 2.708 peserta JKN penderita DM yang tersebar di 141 FKTP dan dianalisis secara berjenjang menggunakan model regresi Multilevel Fractional Logit. Hasil penelitian menunjukkan nilai Intraclass Correlation Coefficient (ICC) sebesar 8,75%, yang berarti 91,25% variasi proporsi pemanfaatan insulin berhubungan dengan faktor internal pasien (Level 1) dan 8,75% berhubungan dengan faktor fasilitas kesehatan (Level 2). Pada level individu, peningkatan umur (p = 0,003; dy/dx = 0,20%) dan status hak kelas rawat Kelas 1 (p = 0,001; dy/dx = 6,63%) secara signifikan meningkatkan persentase pemanfaatan insulin PRB. Sebaliknya, keberadaan komorbiditas menurunkan pemanfaatan insulin PRB di FKTP sebesar 10,60% (p = 0,000; dy/dx = -10,60%) akibat terjadinya pergeseran lokasi layanan utama (shifting of care) ke rumah sakit. Pada level fasilitas kesehatan, ketersediaan jejaring apotek PRB menjadi satu-satunya determinan fasilitas kesehatan yang signifikan meningkatkan proporsi pemanfaatan insulin sebesar 22,38% (p = 0,049; dy/dx = 22,38%). Indikator mutu administratif fasilitas kesehatan seperti Capaian KBK, Angka Kontak, dan nilai KESSAN tidak berhubungan signifikan apabila hambatan pada ketersediaan insulin PRB belum diselesaikan. Kesimpulannya, perluasan kerja sama jejaring apotek PRB di FKTP harus menjadi prioritas utama kebijakan untuk memastikan pemanfaatan insulin PRB.

Diabetes Mellitus (DM) requires continuous, long-term chronic disease management administered through the Chronic Disease Referral Program (Program Rujuk Balik/PRB) at Primary Health Care Facilities (FKTP) to minimize the service burden on hospitals. However, the achievement of PRB insulin utilization at primary health care facilities has not been optimal due to various determinants from both the patient side (demand side) and health facilities (supply side). This study aims to determine the relationship between individual and health facility level determinant factors on the proportion of PRB insulin utilization. This study is an analytical observational study with a cross-sectional approach utilizing secondary data from the Self-Service Business Intelligence (SSBI) application of BPJS Kesehatan in five Regencies/Cities in Jambi Province for the 2024 service period. A final sample of 2,708 National Health Insurance (JKN) participants diagnosed with DM was distributed across 141 FKTPs and analyzed hierarchically using a Multilevel Fractional Logit Regression model. The results demonstrated an Intraclass Correlation Coefficient (ICC) value of 8.75%, which means that 91.25% of the variance in the proportion of insulin utilization was associated with patient-level individual factors (Level 1) and 8.75% was associated with health facility-level characteristics (Level 2). At the individual level, an increase in age (p = 0.003; dy/dx = 0.20%) and a Class 1 premium tier status (p = 0.001; dy/dx = 6.63%) significantly increased the percentage of PRB insulin utilization. Conversely, the presence of comorbidities decreased PRB insulin utilization at FKTP by 10.60% (p = 0.000; dy/dx = -10.60%) due to a shifting of care back to referral hospitals. At the health facility level, the availability of a PRB network pharmacy emerged as the sole significant health facility determinant increasing the proportion of insulin utilization by 22.38% (p = 0.049; dy/dx = 22.38%). Health facility administrative performance and quality metrics such as Performance-Based Capitation (KBK) compliance, contact rates, and customer satisfaction indexes (KESSAN) failed to yield statistical significance when barriers regarding PRB insulin availability remained unresolved. In conclusion, expanding the cooperation of contract PRB pharmacies at FKTP must be the main policy priority to ensure PRB insulin utilization.