Prevalensi karies gigi yang ditunjukkan dengan decayed, missing dan filled teeth (DMF-T) masih merupakan masalah kesehatan masyarakat, termasuk di Indonesia. Penelitian ini bertujuan untuk mengetahui peran faktor individu, rumahtangga dan kabupaten/kota terhadap kejadian karies gigi guna menyusun model pengendalian karies gigi di Kepulauan Bangka Belitung (provinsi dengan riwayat karies gigi tertinggi). Desain penelitian ini campuran (hybrid) antara ecological study dan cross-sectional. Pengumpulan data dilakukan dengan wawancara menggunakan kuesioner, pemeriksaan kesehatan gigi, pengambilan sampel air dan wawancara mendalam. Analisis data menggunakan regresi logistik multilevel (dengan mixed-effect model). Hasil penelitian menunjukkan faktorfaktor pada tingkat individu (frekuensi menggosok gigi, kebersihan gigi dan mulut dan kebiasaan makan makanan asam/bercuka), tingkat rumahtangga (jenis sumber dan keasaman air) dan tingkat kabupaten/kota (ketersediaan perawat gigi dan dokter gigi, angka gizi buruk dan besar anggaran kesehatan per kapita) berpengaruh terhadap prevalensi karies gigi berat pada penduduk dewasa, dimana secara keseluruhan dapat menjelaskan variasi risiko karies gigi sebesar 73,6%. Model pengendalian karies gigi yang sesuai dengan kondisi Kepulauan Bangka Belitung adalah menggabungkan pengendalian faktor pada tingkat individu, rumahtangga dan kabupaten/kota.

---

Dental caries prevalence, indicated by decayed, missing and filled teeth (DMF-T), remains a global public health problem, including Indonesia. The objective of this research was to address the role of individual factors, households, and regency/municipality in explaining dental caries incidence, in order to formulate a model to control dental caries in Kepulauan Bangka Belitung—the province with the highest dental caries prevalence in Indonesia. This research was designed as a combination (hybrid) of cross-sectional and ecological studies. Quantitative and qualitative data were collected through interview, dental health examination by dentists, water sampling, and in-depth interviews. A multilevel logistic regression (mixed-effect) model was fitted to the data. The results show that the explanatory variables at individual (frequency of teeth brushing, dental and mouth hygiene, and acidic food consumption), household (main water source and acidity), and regency/municipality (availability of dentist and dental nurse, malnutrition, and per capita health budget) levels influenced the prevalence of severe dental caries among adults; they all explained 73.6% of the variation in risk of dental caries. The appropriate disease control model, given the local conditions of Kepulauan Bangka Belitung Province, is one that integrates control of risk factors at individual, household, and regency/municipality levels.