<div style="text-align: justify;">Stunting balita di Indonesia tercatat sebesar 19,8% pada Survei Status Gizi Indonesia (SSGI) 2024, turun dari 21,5% pada 2023, namun masih jauh dari target Rencana Pembangunan Jangka Menengah Nasional (RPJMN) sebesar 14,2% pada 2029, terutama pada periode 1000 Hari Pertama Kehidupan (HPK). Penelitian ini bertujuan menyusun model struktural determinan status gizi pada balita usia 6&ndash;23 bulan di Indonesia serta mengidentifikasi prioritas intervensi berdasarkan kontribusi efek total faktor maternal, faktor lingkungan, dan mediator proksimal menggunakan generalized structural equation modeling (GSEM) berbasis data individu SSGI 2024. Desain penelitian adalah potong lintang kuantitatif; dari 68.982 balita usia 6&ndash;23 bulan yang telah dikeluarkan observasi dengan konsumsi TTD dan status ASI bernilai nol (non-exposure), penerapan complete case analysis menghasilkan sampel analisis akhir sebanyak 45.443 balita (65,9%). GSEM diestimasi melalui enam persamaan struktural menggunakan Stata dengan prosedur bootstrap 1.000 replikasi berkelompok pada level klaster (PSU), dengan keluarga distribusi dan fungsi taut disesuaikan pada skala pengukuran tiap variabel endogen. Hasil menunjukkan rata-rata z-skor tinggi badan menurut umur (HAZ) sebesar -1,11, mengindikasikan defisit pertumbuhan linear. Berat lahir bayi, ASI eksklusif, dan skor morbiditas berpengaruh signifikan langsung terhadap z-skor berat badan menurut umur (WAZ), sedangkan Minimum Acceptable Diet (MAD) dan sanitasi tidak menunjukkan pengaruh langsung yang signifikan terhadap WAZ, meskipun skor sanitasi berpengaruh positif signifikan terhadap MAD. WAZ selanjutnya berpengaruh sangat kuat dan sangat signifikan terhadap HAZ (&beta; terstandar=0,6124; R&sup2;=37,50%), menjadikannya mediator akhir paling dominan dalam model. Berdasarkan dekomposisi efek total, jumlah anak menunjukkan efek total terbesar dan signifikan terhadap HAZ (+0,0157), diikuti konsumsi TTD yang signifikan meski berbesaran sangat kecil (+0,0001), sementara usia ibu (-0,0016) dan skor sanitasi (-0,0141) tidak menunjukkan efek total yang signifikan. Uji sensitivitas full versus partial mediation menemukan bahwa berat lahir bayi dan skor morbiditas memiliki jalur pengaruh langsung yang signifikan terhadap HAZ di luar mediasi penuh oleh WAZ (p<0,001), sehingga asumsi mediasi penuh pada kerangka PPCT tidak sepenuhnya didukung. Temuan ini menegaskan pentingnya intervensi konvergen yang memprioritaskan perencanaan jarak kelahiran, kepatuhan konsumsi tablet tambah darah, pengendalian morbiditas, dan pemantauan berat badan dini sebagai titik ungkit penurunan stunting di Indonesia.</div><hr /><div style="text-align: justify;">Childhood stunting in Indonesia stood at 19.8% based on the 2024 Indonesian Nutritional Status Survey (SSGI), down from 21.5% in 2023, yet still far from the National Medium-Term Development Plan (RPJMN) target of 14.2% by 2029, particularly during the first 1,000 days of life. This study aimed to develop a structural model of nutritional status determinants among children aged 6&ndash;23 months in Indonesia and to identify intervention priorities based on the total effect contribution of maternal factors, environmental factors, and proximal mediators, using generalized structural equation modeling (GSEM) based on individual-level data from the 2024 SSGI. A quantitative cross-sectional design was employed; from 68,982 children aged 6&ndash;23 months after excluding observations with zero iron-folic acid (IFA) tablet consumption and zero breastfeeding exposure (non-exposure), complete case analysis yielded a final analytic sample of 45,443 children (65.9%). GSEM was estimated through six structural equations using Stata with a 1,000-replication bootstrap procedure clustered at the primary sampling unit (PSU) level, with the distribution family and link function for each endogenous variable specified according to its measurement scale. Results showed a mean height-for-age z-score (HAZ) of -1.11, indicating a linear growth deficit. Birth weight, exclusive breastfeeding, and morbidity score had significant direct effects on weight-for-age z-score (WAZ), whereas Minimum Acceptable Diet (MAD) and sanitation showed no significant direct effect on WAZ, although sanitation score was significantly and positively associated with MAD. WAZ subsequently exerted a very strong and highly significant effect on HAZ (standardized &beta;=0.6124; R&sup2;=37.50%), establishing it as the most dominant terminal mediator in the model. Based on the total effect decomposition, number of children showed the largest significant total effect on HAZ (+0.0157), followed by IFA consumption, which was significant despite its very small magnitude (+0.0001), while maternal age (-0.0016) and sanitation score (-0.0141) showed no significant total effects. A full-versus-partial mediation sensitivity test found that birth weight and morbidity score retained significant direct paths to HAZ beyond full mediation through WAZ (p<0.001), indicating that the full-mediation assumption of the PPCT framework was not fully supported. These findings underscore the importance of convergent interventions prioritizing birth spacing, iron-folic acid supplementation adherence, morbidity control, and early weight monitoring as leverage points for reducing stunting in Indonesia.</div>